What shows up when people look you up.
Plain-English guides to the record that decides what Google shows, what AI assistants say, and who buyers believe: press placements, podcast footprints, entity signals, and the authority work that ties them together. Every guide connects to the services and pricing behind it.
How-To Guides and Comparisons
Step-by-step procedures built from official documentation, with a check that each step worked and a table of common problems.
How to Write Podcast Show Notes: Summary, Chapters, Guest Credits, Links and Disclosures
Seven steps to show notes for a company or executive podcast: a guest intake form, a summary that fits Spotify's 250-character preview, chapters Apple, Spotify and YouTube can read, podcast:person credits and sponsor disclosures, all checked against each platform's limits.
How to Read Podcast Analytics in Apple Podcasts and Spotify: Plays, Engaged Listeners, Downloads and Why the Numbers Differ
Apple counts a listener after any playback, Spotify a play at 30 seconds, and a certified download needs a minute of audio requested. Seven steps for reading all three reports for one episode, waiting out Apple's 72-hour delay, and comparing against the show's baseline.
What Is Audited Circulation? AAM, ABC, Publisher's Statements and What an Audit Actually Checks
Audited circulation is an audience number checked by an independent body: AAM (founded 1914, merged with BPA in 2023) or the UK's ABC (1931). How a Publisher's Statement differs from an audit report, and what qualified, nonpaid and nonqualified mean on a real statement.
What Is a Podcast Network? Definition, How Networks Make Money and How Rankers Count Them
A podcast network groups shows to sell their ads and often to produce and promote them. Podtrac's definitions, Acast's revenue-share model, real networks from iHeart to NPR, and why Podtrac, Triton and Edison each name a different leader.
What Is Dynamic Ad Insertion? How Podcast Ads Are Stitched In, and What It Means for Guests and Sponsors
Dynamically inserted ads made up 92% of U.S. podcast ad revenue in 2022, up from 48% in 2019, according to IAB and PwC. What the IAB definitions say, how hosting platforms stitch ads into old episodes, how delivery is counted, and what it means for an executive's past interviews.
How to Find a Publication's Editorial Calendar: Where It Lives, What the Deadlines Mean and How to Pitch a Feature
Where publications keep their editorial calendars, and how to read the dates. Ad close is usually not your deadline: at Pharmaceutical Engineering, articles are due 66 days before ads close. Covers pitching a themed issue, outlets with no calendar, and why themes change.
How to Write a Letter to the Editor: Word Limits, Response Windows, Exclusivity and Disclosure
Letters pages run 150 to 400 words, want replies within days and expect exclusivity. Using the NYT, Washington Post, Guardian and local papers' own rules: authorship, AI bans, verification calls, disclosing interests, and when to send a correction request or an op-ed instead.
How to Write a Contributed Article for a Trade Publication: Guidelines, Abstract, Draft, Disclosure and Links
Trade titles publish outside experts on their own terms. Six publications' contributor guidelines: 600 to 2,000 words, first-run or exclusive, AI capped or banned, no product mentions. Then pitching an abstract, disclosure, and what Google says about links in guest posts.
What Is a Masthead? The Staff List, the Front-Page Title and Why PR Teams Check It
A masthead is either a publication's list of editors and staff or, in British usage, its front-page title. Dictionary definitions, real mastheads from Harper's, The Nation and CJR, and how PR teams use one to find the right editor, tell staff from contributors and vet an outlet.
What Is Content Syndication? Wire Services, Licensing, Canonical Tags and What It Means for PR
Content syndication means one story licensed onto many sites, from AP and Reuters copy to Yahoo partner pages. What Google now advises for syndicated copies, how 24 Yahoo-hosted stories declared their canonical URL, and the separate B2B lead-gen meaning of syndication.
301 vs 302 Redirect: Which Is Better for Moving a Page in 2026?
Use a 301 when a page has moved for good and a 302 for a temporary detour. RFC 9110 definitions, how Google reads each as a strong or weak canonical signal, 307 and 308, redirect chain limits, and how long to keep redirects.
Canonical Tag vs Noindex: Which Is Better for Handling Duplicate Pages in 2026?
For duplicate pages, Google calls rel="canonical" the preferred solution; noindex removes a page from results entirely. What each tells Google, what happens to links and signals, crawling, and the robots.txt pitfall.
What Is a Crawl Budget? How Google Defines It and When It Matters
Crawl budget is the set of URLs Google can and wants to crawl on a site. Google's definition, crawl capacity and demand, the site sizes where it matters, what wastes it, the Crawl Stats report, and common myths.
What Is a Robots Meta Tag? The Rules Google and Bing Read, and What They Do
A robots meta tag tells crawlers whether to index a page, follow its links or show a snippet. Google's 12 valid rules and three ignored ones, Bing support, X-Robots-Tag, conflicts, AI Overviews, and why robots.txt blocks hide it.
How to Check Your Site for Broken Links: Search Console, a Crawler and a Fix List
Seven steps to find and fix broken links: Search Console 404 and soft 404 reports, a free crawler, checking which dead URLs still get links from press coverage, and when to update a link, redirect it or return 404 or 410.
How to Add Breadcrumb Structured Data: BreadcrumbList JSON-LD, Visible Trails and Validation
Add breadcrumb structured data in seven steps: map the path a typical visitor follows, show it as accessible links, write BreadcrumbList JSON-LD, then validate and monitor it. Since January 2025, Google shows breadcrumbs in desktop results only.
How to Add Local Business Schema Markup: LocalBusiness JSON-LD, Hours, Location and Validation
Add LocalBusiness schema markup step by step: choose the most specific subtype, include the required name and address, add hours, coordinates and phone in Google's formats, match your Business Profile, then validate.
What Is a Knowledge Graph? Definition, How Google's Works and What It Is Not
A knowledge graph stores facts as links between entities. How Google's Knowledge Graph began, its reported size and sources, how it differs from a knowledge panel and from Wikidata, and what it means for companies.
What Is a Featured Snippet? Definition, How Google Chooses Them and What They Do to Clicks
A featured snippet is a Google result that shows a page excerpt before the link. How Google chooses them, why they can't be marked up, how to opt out, what click studies show and how they relate to AI Overviews.
What Is Zero-Click Search? Definition, the Data and What It Means for Brands
A zero-click search ends without a click on any result. What SparkToro, Pew and Ahrefs data show, Google's counter-claim that clicks are stable, why the studies' figures differ and what it means for brands.
How to Write a Podcast Guest One-Sheet: Bio, Topics, Questions, Proof and a File Hosts Can Open
A seven-step guide to a one-page podcast guest sheet with a verified bio, specific episode topics, sample questions and linked proof, plus keeping the PDF under email attachment limits and disclosing paid appearances under FTC guidance.
How to Create a YouTube Channel for a Podcast: Brand Account, Verification, Podcast Playlist and Episodes
Set up a YouTube channel for a podcast: a Brand Account channel, phone verification to unlock uploads over 15 minutes, a podcast playlist with a square thumbnail, episode titles, chapters, Shorts and the RSS option for audio-only shows.
Riverside vs Zencastr: Which Is Better for Remote Podcast Recording in 2026?
Riverside vs Zencastr compared from vendor pages in October 2026: free plans, paid tiers, participant limits, recording quality and team features, with a verdict on which suits which kind of show.
LinkedIn Newsletter vs Substack: Which Is Better for a Professional Newsletter in 2026?
LinkedIn newsletters vs Substack, compared from official pages in October 2026: reach and notifications on LinkedIn, Substack's 10% fee on paid subscriptions, list ownership and export. Verdict: LinkedIn for reach, Substack for ownership.
How to Create a LinkedIn Newsletter: Eligibility, Setup, First Edition and Subscribers
A seven-step guide to creating a LinkedIn newsletter as a member or a company Page: access rules, title, description and cadence, recommended image sizes, the first edition, and the five-newsletter limit.
Brand24 vs Mention: Which Is Better for Brand Monitoring in 2026?
Brand24 publishes its plans and mention caps; Mention, now owned by Agorapulse, sells a Company plan with no price shown. Compares pricing, limits, sources, analytics and who should choose each.
EIN Presswire vs PR Newswire: Which Is Better for Distributing a Press Release in 2026?
EIN Presswire publishes per-release packages with stated word, image and link limits; PR Newswire claims a far larger network but publishes no price. Compares cost, limits, editorial rules, links and fit.
Google Trends vs Google Search Console: Which Should You Use to Measure Search Interest and Coverage Impact?
Google Trends shows relative interest in any search term on a 0 to 100 scale; Search Console counts your own site's clicks and impressions. Compares data, filters, limits, freshness and how PR teams should use each.
How to Run a Survey for Press Coverage: Question, Sample, Fieldwork and the Disclosure Journalists Expect
A seven-step guide to commissioning a survey journalists will report: defining the population, choosing and disclosing the sample, sizing for subgroups, screening out bogus respondents, and publishing the method under AAPOR and British Polling Council standards.
How to Set Up a LinkedIn Company Page: Eligibility, URL, Images, Admins and First Followers
A seven-step guide to creating a free LinkedIn Company Page from LinkedIn Help: eligibility rules, URL format, required fields, image sizes, admin roles, and how long employee association takes.
Press Release vs Pitch Email: Which Should You Send, and When?
Press release vs pitch email compared on purpose, length, cost, journalist survey data, CAN-SPAM and Regulation FD. A pitch wins for getting one reporter to write a story; a release wins as the public record. Most announcements need both.
What Is a News Hook? Definition, Types and How Editors Judge One
A news hook is the timely reason a story should run now. The definition, the news values behind hooks from a peer-reviewed study of UK lead stories, common hook types, how editors meet them, and a checklist before you pitch.
What Is a Lede? Definition, Types and Why It Is Spelled That Way
A lede is the opening of a news story that states the main news. The dictionary definition, why it is spelled lede, the main lede types, published length guidance, and how to write one for a press release.
What Is a Dateline? Definition, Format and the Rules Behind It
A dateline states where, and in a press release when, a story comes from. The definition, AP-style format, the AP's ethics rules on datelines, press release examples, and how to keep online dates consistent for Google.
What Is a Media Advisory? Definition, Contents and Timing
A media advisory is a one-page invitation for journalists to cover an event. What it contains, when to send it, with timing from real NASA and Canadian Space Agency advisories, how to distribute it, and how it differs from a press release.
How to Run a Media Training Session for Executives: Ground Rules, Messages, Mock Interviews and Debrief
A step-by-step plan for training executives before interviews: set one objective, teach the on-the-record default and other ground rules, build messages with proof, clear disclosure limits with legal and IR, rehearse recorded mock interviews, and debrief in writing.
Press Kit vs Media Kit: Which Document Do You Need, and Who Is It For?
A press kit gives journalists free facts, bios, images and a contact so coverage is accurate; a publisher's media kit sells advertising with audience data, ad specs and rates. Compares contents, price, tone, FTC and SPJ rules, and when to use each.
How to Write an Award Entry: Category, Evidence, Word Limits and Judging Criteria
Eight steps to an award entry judges can score: choose a programme you can defend, mirror the category criteria, stay inside the eligibility period and word limit, package evidence judges can open fast, budget for rising fees, and report the result accurately.
How to Submit a Conference Speaking Proposal: Format, Title, Abstract, Outline and Follow-Up
How to write a call-for-proposals submission reviewers accept: read the CFP rules, choose a format, write a specific title, an abstract and a timed outline, remove sales pitches, add a past recording, and follow up whether accepted or not.
What Is a Trade Publication? Definition, How It Is Funded and What It Is Not
A trade publication is a newspaper, magazine, newsletter or site produced for people working in one industry. The definition, how trade titles earn money, publisher-stated audience figures, FTC sponsored-content rules, and how companies earn trade coverage.
On Background vs Off the Record: Which Ground Rule Should a Spokesperson Use?
On background lets a reporter publish what you say under an agreed description; off the record means nothing can be published. Compares the AP, NPR and NYU definitions, how to set ground rules, and when a spokesperson should use each.
What Is an Exclusive in PR? Definition, Types and How It Differs From an Embargo
An exclusive gives one outlet a story before, or instead of, any other. The main types of exclusive, how it differs from an embargo, how to offer one with a decision deadline, and the Regulation FD question for listed companies.
How to Plan an Embargoed Press Announcement: Lift Time, Agreement, Package and Breaks
Plan an embargoed announcement in seven steps: decide whether an embargo fits, set one lift time and time zone, get each journalist's agreement before sending, prepare a full package, and decide in advance to lift for everyone if it breaks.
How to Write a Crisis Holding Statement: What to Say First, a Template and the Legal Clocks
Write a crisis holding statement in seven steps, built on CDC crisis-communication guidance: what is known and unknown, empathy, confirmed facts, a few actions and a next-update time. Includes a template and the GDPR, NIS2, SEC and HIPAA deadlines that may apply.
How to Write a Company Boilerplate: Facts, Length, Platform Versions and a Template
Write a company boilerplate in seven steps: approved facts, a first sentence that stands alone, checkable evidence, a tight length, and versions that fit each platform. Includes four real boilerplates measured and a template.
How to Respond to a Negative News Article: Triage, Correction, Statement and Follow-Up
How to respond to unfavourable coverage: classify each claim as error, missing context or fair criticism, request corrections with evidence, publish a short factual statement, and escalate only through routes that fit. Draws on the Editors' Code, SPJ code and Google's removal pages.
How to Get Quoted as an Expert Source: Profile, Requests, Responses and Follow-Up
Seven steps to become a quoted expert: choose narrow topics, build a source profile page, monitor reporter requests, screen for fit, write a short quotable reply in your own voice, follow up and log results. Covers HARO, Qwoted and Featured rules and why not to pay for quotes.
How to Write and Place an Op-Ed: Outlet, Pitch, Draft, Disclosure and Submission
How to pick an outlet, build a one-argument piece, pitch or submit in the format the outlet asks for, and disclose interests. Uses Guardian US, Baltimore Sun and HBR guidelines for lengths and rules.
How to Add Video Structured Data: VideoObject Markup, Testing and Search Console Monitoring
How to add VideoObject JSON-LD to a watch page: required and recommended properties, a worked example, testing in the Rich Results Test and URL Inspection, and monitoring in Search Console. Notes that Google does not guarantee a video result.
How to Fix a Wrong Fact on Wikipedia About Your Company: Disclosure, Sources and the Edit Request
How a company employee or agency can correct a wrong fact on English Wikipedia: disclose the connection, find an independent reliable source, post an edit COI request on the talk page, and escalate correctly. Based on Wikipedia's COI, paid-disclosure, verifiability and edit-request pages.
How to Set Up Bing Places for Business: Prepare, Claim or Add, Verify and Check Your Listing
Six steps to claim or add a free Bing Places for Business listing: prepare consistent details, sign in, search for an existing listing, claim or add, verify and check the result in Bing. Written conservatively because Microsoft's public help is thin.
How to Prepare an Executive for a Podcast Interview: From Claims Sheet to Published Episode
Eight steps to get an executive ready for a podcast interview: a claims sheet, a legal and compliance check, a mock interview, a prep pack for the host, recording setup and a post-publication check. Quotes Regulation FD, the FTC endorsement FAQ and Apple Podcasts documentation.
What Is Earned Media? Definition, What Counts and What Does Not
Earned media is coverage a third party chooses to publish without payment for the placement. Quotes Forrester's 2009 definition and the PESO model, separates what counts from wire releases and sponsored posts, and sets out the FTC and Google link rules.
What Is a Backlink? Google's Definition, Uses and Limits
A backlink is a link on another site that points to a page on yours. Quotes Google's Search Console definition, what Google says about links and quality, when paid or press release links become spam, and how to check backlinks in the Links report.
What Is Entity SEO? How Search Engines Identify Companies and People
Entity SEO is an industry term for helping search engines identify which person, company or thing a site is about. The guide shows what Google actually documents about entities and knowledge panels, what it does not, and practical checks for a company or executive.
What Is the sameAs Property? What Schema.org and Google Say About Identity URLs
The sameAs property is a schema.org property whose value is a URL of a page that identifies the same entity. The guide quotes schema.org and Google's Organization documentation, lists which URLs fit, and states that Google documents no ranking effect.
What Is a Rich Result? What Google Says Structured Data Can and Cannot Do
A rich result is a richer Google Search appearance that structured data can make a page eligible for. The guide quotes Google's definition, why eligibility is not a guarantee, which features matter for PR teams and how to test and monitor them.
What Is an Op-Ed? How It Differs From an Editorial and a Sponsored Post
An op-ed is a signed opinion piece by an outside contributor, published on a newspaper or magazine opinion page. This definition covers the origin of the term, how it differs from editorials, columns and sponsored posts, and what Google and the FTC document about paid links and disclosure.
Ahrefs vs Moz: Which Is Better for Link Research and AI Visibility in 2026?
Ahrefs and Moz Pro compared on October 2026 published pricing, plan limits, backlink data claims, Domain Rating versus Domain Authority, AI visibility tracking and API access. Ahrefs suits data-heavy link research; Moz is cheaper to start.
Prowly vs Muck Rack: Which Is Better for Media Outreach in 2026?
Prowly has become the Semrush AI PR Toolkit, so this comparison sets it against Muck Rack on published pricing, plan limits, journalist databases, pitching and monitoring. The toolkit starts at $149 a month billed annually; Muck Rack does not publish a price.
Earned Media vs Paid Media: Which Is Better for Brand Credibility in 2026?
Earned media is coverage a third party chooses to give you; paid media is a placement you buy. This comparison sets them against each other on credibility, control, disclosure, links and measurement, using FTC and Google documentation, and covers the grey zone of sponsored content.
Schema.org vs Open Graph: Which Do You Need for Search and Sharing in 2026?
Schema.org describes what a page is about for search engines; Open Graph controls how a shared link previews on platforms. This comparison sets out purpose, syntax, required fields and what Google documents, and concludes that most sites need both.
Podcast Guesting vs Podcast Advertising: Which Is Better for Authority in 2026?
Podcast guesting is normally unpaid and earned; podcast advertising is paid and controlled. This comparison sets them against each other on credibility, control, disclosure, links and measurement, and warns that a paid guest spot counts as advertising.
Not llms.txt, But Crawlable Pages: What AI Search Actually Reads
Google says you don't need new AI text files to appear in AI Overviews, and the crawler docs of OpenAI, Anthropic and Perplexity describe robots.txt-controlled bots that read ordinary pages. What the evidence shows about llms.txt, where it still helps, and the crawlability checks that matter more.
Not More Press Releases, But Earned Coverage: What Google and Journalists Actually Reward
Google's spam policy names optimised press-release links as link spam, and wire body links in our checks were nofollow. What press releases are still good for, and how to shift effort toward coverage journalists choose to write.
Not a Knowledge Panel, But a Consistent Entity: What Google Actually Lets You Control
Google says it does not manually create knowledge panels and that they are generated from sources across the web. What claiming does and does not let you change, and how consistent facts, Organization markup with sameAs and corrected sources give Google something reliable to work with.
How to Pitch a Journalist by Email: Subject Line, Length, Follow-Up and the Rules That Apply
A step-by-step method for pitching one journalist by email: choosing the right reporter, an accurate short subject line, a short body and one follow-up. Includes a template, labelled vendor survey data, and what CAN-SPAM, the UK ICO and the SPJ Code mean for outreach.
How to Create an Author Page: Bylines, Bio, ProfilePage Markup and Article Author Links
How to build an author page Google and readers can use: one stable URL per real author, a factual bio, bylines that link to it, ProfilePage markup and Article author.url pointing at it, with Google's own author markup rules.
How to Get Your Company on Crunchbase: Create, Verify and Complete the Profile
How to list a company on Crunchbase from its own help center: registering, creating or claiming the profile, and completing it to Crunchbase's content rules. Covers moderation, fields only staff can change, and how Wikipedia treats Crunchbase as a source.
How to Get Into Google Discover: Eligibility, Images, Content Policies and the Discover Report
There is no way to submit a site to Google Discover; this guide covers what Google says makes content eligible and more likely to appear: indexing, Discover's content policies, large images with max-image-preview:large, bylines and dates, and the Search Console Discover report.
Cision vs Meltwater: Which Is Better for Media Monitoring and PR in 2026?
Cision and Meltwater compared from their own pages, checked October 2026: monitoring coverage, social data, API access, support, contract terms and ownership. Neither publishes prices.
Google Business Profile vs Apple Business: Which Is Better for Local Listings in 2026?
Google Business Profile and Apple Business compared from each company's own help documentation: where listings appear, reviews, posts, insights, ads and verification. Both are free, and most local businesses should claim both.
Buzzsprout vs Spotify for Creators: Which Is Better for Podcast Hosting in 2026?
Buzzsprout and Spotify for Creators compared for podcast hosting from their own pricing and help pages: cost, distribution to Apple, Spotify and YouTube, video, analytics, monetization and leaving with your feed.
How to Build a Media List: Find the Right Journalists, Record Why Each Is There and Keep It Current
A seven-step method for building a media list around one story: choosing outlets, finding the journalist who covers the beat, verifying roles and recording why each contact is there. Covers list fields, segmentation, opt-outs, and what the FTC and UK ICO say about emailing business contacts.
How to Write a Press Release: Structure, Wire Rules and a Fill-In Template
An eight-step guide to writing a press release: news angle, headline, dateline and lead, supporting facts, quotes, links, boilerplate and media contact, with a fill-in template. Draws on wire services' published editorial rules, Google's link spam policy and Regulation FD for public companies.
How to Create an Online Newsroom: Release Pages, Media Assets, Markup and Disclosure
How to set up a newsroom on your own website in seven steps: stable URL, dated release pages, media contact, company facts, licensed images, and sitemap and structured data. Includes what the SEC says about using a company website to disclose information.
How to Write an Executive Bio for Media: Three Lengths, Verified Facts and a Template
How to write a media-ready executive bio in three lengths, about 50, 150 and 350 words, with every fact verified and one canonical bio page on your own site. Covers what Google says about authorship and fabricated credentials, and the SEC's Item 401 bio rule for public companies.
What Is a Press Kit? What It Contains, What It Is Not and How It Differs From a Newsroom
What a press kit is, quoted from Merriam-Webster, and what it typically contains: release, fact sheet, bios, images and a media contact. Explains how it differs from an online newsroom and an advertising media kit, and what the SPJ code and Google's spam policies say about gifts and samples.
How to Verify Your Site in Bing Webmaster Tools: Search Console Import, XML File, Meta Tag and DNS
How to add and verify a site in Bing Webmaster Tools, from Bing's own help: importing from Google Search Console, DNS auto verification, the BingSiteAuth.xml file, the meta tag and a CNAME record. Covers account limits, user roles and what to do when verification fails.
How to Fix “Crawled – Currently Not Indexed” in Google Search Console: 7 Steps
How to diagnose and fix “Crawled – currently not indexed” in Google Search Console, using Google's Page indexing and URL Inspection documentation. Covers technical lookalikes (soft 404s, unrendered JavaScript, near-duplicates), which pages to keep, merge or remove, and when requesting indexing helps.
How to Track Branded Search in Google Search Console: The Branded Filter, Regex and Monthly Reporting
How to measure branded and non-branded search in Google Search Console using the Branded query filter, the Insights card and a regex brand filter. Covers availability limits, anonymized queries, period comparisons and how to report branded share monthly.
What Is a 301 Redirect? How HTTP Defines It and How Google and Bing Treat It
What a 301 redirect is, from RFC 9110's definition of “Moved Permanently”, and how Google and Bing treat it. Compares 301 with 302, 307 and 308, covers chains, hop limits and how long to keep redirects, and lists common misconceptions with the source for each.
Noindex vs robots.txt Disallow: Which Is Better for Keeping Pages Out of Search in 2026?
A comparison of noindex and robots.txt Disallow, built from Google and Bing documentation and RFC 9309. Explains why noindex is the tool for keeping pages out of results, why Disallow is for managing crawling, and why using both on one URL defeats the noindex.
How to Claim Your Apple Business Connect Listing: Claiming Your Apple Maps Place Card in Apple Business
Apple Business Connect became Apple Business on April 14, 2026, and Apple Maps place cards are now claimed there. The steps from Apple's guide: finding your place, verification, transfer requests, place card content and Insights.
What Are Core Web Vitals? LCP, INP and CLS, and How Google Uses Them in 2026
The three Core Web Vitals (LCP, INP and CLS) and their good, needs-improvement and poor thresholds, quoted from web.dev and Google Search Central. Covers INP replacing FID, how field data is measured and what Google says the metrics do and do not do for ranking.
What Is hreflang? How Google Uses It to Serve the Right Language or Regional Page
What hreflang is and how Google uses it, from Google's own documentation. Covers the three equivalent methods, return links, x-default, which language and region codes are valid and how hreflang differs from rel="canonical".
What Is Anchor Text? What Google Says Good Link Text Looks Like
Google's definition of anchor text and its examples of good and bad link text, from Search Central's link guidance. Covers the title and alt-text fallbacks, the spam policy on optimized anchor text in press releases and guest posts, and Search Console's Top linking text report.
Google AI Overviews vs AI Mode: Which Matters More for Your Site’s Visibility in 2026?
How Google's two generative AI search features differ for site owners: where each appears, how it links to sources, how Search Console counts it and what you can measure. Both share the same eligibility rules and opt-out controls.
What Is the PESO Model? Paid, Earned, Shared and Owned Media Explained
What the PESO Model is, in Gini Dietrich's own words from Spin Sucks: paid, earned, shared and owned media planned as one system. Covers where it came from, how its category lists changed, and how to classify wire releases, paid placements and podcast appearances.
What Is Share of Voice? Definitions, Formula and How to Measure It
What share of voice means in advertising, PR, search and AI answers, with each definition quoted from its source. Shows the formula with a worked example, explains excess share of voice from Binet and Field's research and its caveats, and sets out what the metric does not measure.
Semrush vs Ahrefs: Which Is Better for Backlink and AI Visibility Research in 2026?
Semrush and Ahrefs compared from their own pricing and product pages, checked October 2026. Covers plan prices and limits, backlink and keyword index claims, AI visibility tracking, mention monitoring, API access, free tiers, and Adobe's acquisition of Semrush.
Perplexity vs ChatGPT Search: Which Is Better for Getting Your Company Cited in 2026?
Perplexity and ChatGPT search compared for companies that want to be cited, using only each firm's own documentation. Covers the crawlers to allow, which fetchers may ignore robots.txt, what blocked pages can still show, source labels, publisher programs and ChatGPT's utm_source tag.
PR Newswire vs GlobeNewswire: Which Is Better for Distributing a Company Press Release in 2026?
PR Newswire and GlobeNewswire compared from both vendors' sites, ownership filings and our own check of 70 GlobeNewswire releases. Covers who owns each wire in October 2026, unpublished pricing, filing services, verification, AI-citation reporting and nofollowed links.
How to Add Person Schema for Executives and Authors: ProfilePage, Author Markup and a JSON-LD Template
Seven steps and a ProfilePage JSON-LD template for describing an executive or author with Person markup, built from Google's ProfilePage and author documentation. Google guarantees no rich result, and the markup does not create a knowledge panel.
How to Control How Google Shows Your Content in AI Overviews and Snippets: nosnippet, max-snippet and data-nosnippet
Page-level controls (nosnippet, max-snippet, data-nosnippet, noindex) limit what Google shows in snippets and AI features alike, while Search Console's Search generative AI control opts a whole property out of AI Overviews and AI Mode. What each does, what Bing documents, and what you give up.
How to Request a Wikipedia Article About Your Company: Notability, Disclosure and Articles for Creation
Eight steps through Wikipedia's notability guideline for companies, conflict-of-interest and paid-editing disclosure rules, Articles for Creation and edit requests, with each page's status labelled. No route guarantees an article.
How to Move Your Website to a New Domain and Protect Search Traffic: Redirects, Change of Address and Bing
Ten steps for a domain move: URL mapping, 301 redirects, updated canonicals and sitemaps, Search Console's Change of Address tool and its requirements, what Bing documents, and how long Google says to keep redirects.
How to Track Press and Podcast Links With UTM Parameters in GA4: Builder, Naming and Reports
Eight steps to tag press and podcast links with UTM parameters, name them so GA4 sorts them into the right channel, and read the results, including why a medium of podcast or pr lands in Unassigned.
What Is YMYL? Google's "Your Money or Your Life" Topics Explained
What “Your Money or Your Life” means, quoted from Google's Search Quality Rater Guidelines: the four harm categories, the spectrum from clear YMYL to not YMYL, the business topics the guidelines name, and Google's statement that raters have no control over rankings.
What Is a Nofollow Link? rel="nofollow", "sponsored" and "ugc" Explained
What rel=nofollow, rel=sponsored and rel=ugc mean, quoted from Google's documentation: the 2019 switch to treating them as hints, the link spam policy on paid links, what Bing documents, and how to check a link's rel value yourself.
What Is a Google AI Overview? What Google Says It Is, When It Appears and How It Links
Google's own definition of AI Overviews: when they appear, how they link, what makes a page eligible, how they show up in Search Console, how AI Mode differs, and the site-level opt-out Google added in 2026.
What Is an XML Sitemap? The Protocol, Its Limits and What It Does Not Do
An XML sitemap lists the URLs a site wants search engines to know about. The protocol, its 50,000-URL and 50MB limits, index files, Google's and Bing's guidance on lastmod, the image, video and news extensions, and what a sitemap does not do.
Google Alerts vs Talkwalker Alerts: Which Is Better for Free Brand Monitoring in 2026?
Google Alerts is the better free default for most brands; Talkwalker Alerts adds X mentions, a published Boolean syntax, delivery to any email address and Slack. Sources, languages, terms and ownership compared from each tool's own pages.
Meltwater vs Muck Rack: Which Is Better for Media Monitoring and Journalist Outreach in 2026?
Muck Rack is the better default for outreach-led PR teams; Meltwater wins when monitoring is the bigger job, with broadcast in all 210 US markets by its own count. Neither publishes prices, and Meltwater's standard term is 36 months.
What Is E-E-A-T? What Google's Rater Guidelines and Search Docs Actually Say
E-E-A-T stands for Experience, Expertise, Authoritativeness and Trustworthiness, the qualities Google's search quality raters are told to look for. Google says E-E-A-T itself isn't a specific ranking factor; this explains where it comes from, what Trust means and what a company can actually do.
What Is a Canonical URL? How Google Chooses One and How rel="canonical" Works
A canonical URL is the version of a page Google picks to represent a set of duplicates. How Google chooses it, how strong redirects, rel=canonical and sitemaps are as signals, what Bing says, and the common mistakes that split one page into two addresses.
How to See Which Sites Link to You in Google Search Console: The Links Report, Exports and Press Checks
Eight steps to read Search Console's Links report and Bing's Backlinks report, confirm a press placement actually links to you, and check the link's rel attribute yourself, since Google's report is a sample and doesn't show nofollow.
How to Set Up a Google Business Profile: Eligibility, Verification and the Rules That Keep It Live
Ten steps to create or claim a Google Business Profile, from Google's eligibility rule (in-person contact with customers) through verification, name and category rules, hours, service areas, photos and managers. Online-only companies are not eligible.
Muck Rack vs Cision: Which Is Better for Earned-Media Outreach and Monitoring in 2026?
Muck Rack suits teams whose main job is pitching journalists; Cision's CisionOne suits teams that mostly need monitoring, especially broadcast, non-US and social, or want media data, PR Newswire and Brandwatch from one owner. Neither publishes prices.
How to Monitor Brand Mentions With Google Alerts and Free Tools: Setup, Operators and Limits
Eight steps to track your company's name with Google Alerts, search operators, Google News and Search Console's branded-queries filter, plus what these free tools cannot see, such as AI answers and Reddit threads.
How to Report a Scam Ad Impersonating Your Company: Google, Meta, Microsoft, X, TikTok, LinkedIn, the FTC and IC3
Nine steps to report an ad that impersonates your company on Google, Meta, Microsoft, X, TikTok and LinkedIn, using each platform's own forms and conditions, then report the landing page as phishing, file with the FTC and IC3, and warn customers.
How to Verify Your Site in Google Search Console: DNS, HTML File, Meta Tag and Every Other Method
Every Google-documented way to verify a site in Search Console: DNS TXT or CNAME, HTML file, meta tag, Google Analytics and Tag Manager. Also covers Domain vs URL-prefix properties, adding owners, what happens when verification lapses, and importing the site into Bing Webmaster Tools.
How to Check If a Page Is Indexed by Google: URL Inspection, Page Indexing and the Limits of site:
How to confirm whether Google has indexed a page using Search Console's URL Inspection tool and Page indexing report, and what each status means. Explains, from Google's own documentation, why a site: search can confirm a page is indexed but cannot prove it is missing.
How to Remove Outdated Content from Google Search: Refresh, Removals and Permanent Fixes
Google only reflects the live page. Seven steps covering the Refresh Outdated Content tool for pages you don't own, the Search Console Removals tool and 404, 410 or noindex for your own, and Google's personal-information and legal request routes.
How to Add Article Structured Data to News and Blog Posts: JSON-LD, Authors and Validation
Seven steps and a JSON-LD template for Article, NewsArticle or BlogPosting, built from Google's documentation on headlines, images, dates and authors. Google lists no required properties, guarantees no rich result and does not say the markup affects ranking.
How to Set Up IndexNow on Your Website: Key File, Submissions and Bing Verification
Generate a key, host it at your site root, then send changed URLs by GET or in batches of up to 10,000 by POST. IndexNow reaches Bing, Yandex, Naver, Seznam.cz, Yep and Amazon, not Google.
How to Add Transcripts to Your Podcast: Apple, Spotify, YouTube, RSS and Your Website
Create a timed VTT file, link it with the podcast:transcript RSS tag and set Apple to display transcripts you provide, then check Spotify's, caption the YouTube video and publish the text on your own site.
How to Get Verified on LinkedIn: Identity, Workplace and Company Page Verification
Seven steps to add LinkedIn identity and workplace verification through CLEAR, Persona, DigiLocker or a work email, and to verify a company Page with a DNS record. Covers which method applies by country, Page eligibility, and what LinkedIn and its partners keep.
How to Set Up a Publication in Google News Publisher Center: What Still Works in 2026
Google News no longer takes submissions: since 2024 and 2025 it builds publication pages automatically and ignores feeds and sections set in Publisher Center. Eight steps covering policy and transparency checks, site name and favicon, Search Console verification, and what Publisher Center still handles.
Google Search Console vs Bing Webmaster Tools: Which Is Better for Monitoring Search and AI Visibility in 2026?
Both are free. Search Console is the main tool for Google indexing and reports AI Overviews and AI Mode impressions only. Bing Webmaster Tools adds Copilot citation data, keyword research, backlink comparison, a site audit and IndexNow.
PR Newswire vs Business Wire: Which Is Better for Distributing a Company Press Release in 2026?
Business Wire publishes a $475 starting price and states UK, Canadian and European disclosure credentials; PR Newswire publishes no price and bundles placement, social and AI reporting. Both mark body links nofollow, and neither can guarantee an AI citation.
Qwoted vs Featured: Which Is Better for Answering Journalist Requests in 2026?
Qwoted limits free users to two pitches a month and bans AI-written pitches. Featured, which owns HARO, keeps HARO's email free and sells AI help from $39 a month, alongside paid backlinks in the same app.
How to Claim and Update Your Google Knowledge Panel: Verification, Suggested Edits and What Google Will Not Change
Seven steps to claim a knowledge panel Google already shows, verify through a profile Google lists, add users and suggest edits. Claiming does not create a panel: Google says it does not manually create or delete them.
How to Create a Wikidata Item for Your Company: Notability, Disclosure and Sourced Statements
Eight steps from the notability test to sourced statements and identifiers, with the paid-editing disclosure the Wikimedia Terms of Use require. Wikidata's self-promotion essay strongly discourages items about your own company, and items that fail notability are deleted.
How to Fix “Discovered – Currently Not Indexed” in Google Search Console: 7 Steps
Google found the URLs but put off crawling them, usually to avoid overloading the site or because demand for them is low. Seven steps: size the problem, inspect samples, cut unwanted URLs, check server health, add internal links, clean the sitemap and request indexing sparingly.
How to Submit a Sitemap to Google and Bing: Search Console, Bing Webmaster Tools and IndexNow
List the sitemap in robots.txt, submit it in Search Console and Bing Webmaster Tools, and add IndexNow for changed URLs. The ping URLs both search engines once accepted are retired: in our check Google's returned 404 and Bing's 410.
How to Block or Allow AI Crawlers in robots.txt: Separate Training From Search
OpenAI, Anthropic, Google and Apple each document a separate robots.txt token for model training, so you can refuse training and stay in AI search. Four copy-paste rule sets, a token reference table, and the user fetchers whose operators say robots.txt may not apply.
How to Create an llms.txt File: Format, Placement and What It Will Not Do
The llmstxt.org format in six steps: an H1, a blockquote summary and H2 lists of annotated links, at the site root or a subpath. Google says AI text files are not needed for its AI features, and no crawler documentation we checked says it reads one.
How to Add Organization Schema Markup to Your Website: A JSON-LD Template and Validation Steps
One JSON-LD block on your home or About page, eight steps, and a copy-paste template. Google says the markup can influence the logo it shows but guarantees no feature, and nothing published links it to AI citations.
How to Track AI Assistant Referral Traffic in Google Analytics 4: Default Channel, Custom Channel Group and Regex
GA4's AI Assistant channel catches recognised chatbot referrals with no setup; a custom channel group regex covers the rest. Google's own example regex over-matches, and visits with no referrer still land in Direct.
How to Submit a Podcast to Apple Podcasts, Spotify and YouTube: One RSS Feed, Three Routes
One RSS feed can be submitted to Apple Podcasts, Spotify and YouTube. Each platform's account, artwork and tag requirements from its own documentation, eight submission steps, and what each says (and does not say) about review time.
How to Pitch Yourself as a Podcast Guest: From Shortlist to Published Episode
Seven steps from choosing and researching shows to the one-sheet, the pitch, the follow-up and the checks to run after the episode airs. Built on Edison Research audience data and platform documentation; no independent survey of pitch response rates exists, and we quote none.
How to Request a Correction to a News Article: A Seven-Step Guide for Companies
Seven steps to getting a factual error corrected: separate fact from opinion, gather evidence, find the right contact, send a request (template included) and escalate to a readers' editor or a regulator such as IPSO. Covers the published policies of the Guardian, BBC, New York Times, AP and Reuters.
How to Check What ChatGPT, Perplexity and Gemini Say About Your Company: A Repeatable Audit
An eight-step audit of what ChatGPT, Perplexity, Gemini and Google's AI Overviews say about a company: a fixed prompt set, recorded test conditions, a logging table, accuracy checks and tracing wrong claims to their sources. Answers vary between runs and accounts, and the method allows for it.
Research
Data reports on podcast and media placement, built from named sources with their type labelled. Where the evidence cuts against paid placement, it is reported as found.
Is a Press Release on USA Today's Site Coverage by USA Today? We Checked 188 From One Week. Google News Listed 48, Every One Under the Paper's Name.
USA Today and Gannett's local papers host paid press releases under a /press-release/ address, and distributors sell placement there by name. We searched Google News for 188 such pages published between 28 September and 5 October. It listed 48 of them, all as “USA Today”, all as the top result, none marked as a release. The address is the only sign, and it is not coverage.
Who Picks the Companies on Forbes, TIME and Fortune Lists? We Sorted 66 Announcements From One Week and Read 18. Most Lists Came From Research Firms, and One Release Mentioned Money.
Companies announce places on magazine lists almost daily. We collected 66 such wire announcements from one week and read 18 in full. Fifteen of the 18 announced lists compiled by an outside firm (Statista, Great Place To Work, Plant-A Insights or SHOOK Research), and 14 named it. Only one said anything about fees, and none said whether the company paid to use the list's name.
What Did Dyson Say as Its $500 Toothbrush Failed? Four Statements in Three Weeks Moved From “Strong Demand” to an Apology. The Cause Came Last.
Dyson's CameraJet went on sale in the US on 8 September and was gone from shops in several countries within a week. We read the company's statements as reported by WIRED, Consumer Reports, Mashable and BeautyMatter. The explanation changed three times, the word recall was used only to deny one, and an apology arrived four weeks after launch.
Does India Now Require AI-Generated Ads to Be Labelled? Only When the AI Could Mislead. A Paid Chatbot Recommendation Must Say “Sponsored by”.
India's advertising self-regulator published final rules on 29 September for labelling synthetic content in ads, taking effect three months later. We read the guidelines and compared them with May's draft. AI-written copy needs no label; a synthetic influencer or a cloned executive voice does; a fabricated testimonial is banned even with one. The named example of a fake AI doctor was dropped.
What Do McClatchy's Cuts Mean for Pitching a Local Paper? Some Newsrooms Lost Half Their Staff in a Day. The Company Is Hiring Editors for AI-Assisted Lifestyle Brands.
McClatchy laid off more than 90 journalists across at least 13 papers on 10 September, cutting city hall, schools and accountability beats. We read the reporting from Poynter, CJR, Straight Arrow News and the papers' own regions. For a company pitching local news, the reporter on the list may be gone, and the company's growth bets are national lifestyle brands and an AI tool that repackages stories.
Can a Company Sue Google Over AI Overviews? A Federal Judge Threw Out Two Antitrust Cases. He Took the Publishers' Losses as True and Said Any Fix Belongs to Congress.
On 30 September Judge Amit Mehta dismissed Chegg's and Penske Media's antitrust suits over AI Overviews in a single opinion. We read all 41 pages. The publishers lost on how they framed the case, not on whether their traffic fell: the court took that as true, and said the fix, if there is one, belongs to Congress.
Do AI Overviews Say What Their Sources Say? An Audit of 55,393 Searches Found 11 Per Cent of Claims Unsupported by the Pages Cited. Three in Ten Cited Sites Were Not on Page One.
Researchers at Washington University in St. Louis captured 7,583 Google AI Overviews, split them into 98,020 factual claims and checked each against the pages Google cited. Most claims held up. The failures were mostly omissions, the cited sources were more credible than the page of results overall, and almost a third came from sites the results page did not show.
Will Text Written With ChatGPT Carry a Watermark? In the EU, Within Weeks. OpenAI's Own Tests Show That Light Editing Hides Most of It.
OpenAI said on 5 October that it will add an invisible watermark to ChatGPT and Codex text in the European Union, to meet the AI Act's marking rule. We read the announcement and the technical report. Only approved researchers will get the detector, short or edited text often escapes it, and a detection says nothing about who wrote what.
Do AI Crawlers Read Your Sitemap? Google's John Mueller Says He Has Seen Them Do It. Of Eight AI Crawler Operators, Only Common Crawl Documents It. 27 of the Dow 30 Point to One.
On Google's Search Off the Record podcast, John Mueller said AI training crawlers have no console to submit a sitemap to, so a site should keep the standard name or rely on RSS. We read eight AI crawler operators' documentation and checked the robots.txt files of the 30 Dow Jones companies to see what they publish.
Did Shopify Put Your Store Into ChatGPT, Gemini, Copilot and Meta? If It Is Eligible, Yes, by Default. Eleven of Fourteen Shopify Stores We Checked Also Serve a File Asking AI Agents to Recommend Shopify's Own App.
Shopify's agentic storefronts share eligible merchants' products with AI channels automatically and enrol them in new ones as they launch, Meta's Muse included. We read Shopify's help pages and the changelog, and checked what fourteen Shopify stores now serve to AI agents at /agents.md.
Which Podcasts Actually Work for B2B Advertisers? An Agency's Ranking Puts a Daily News Show First. On Apple's Own Labels, Four of Its Fifteen Are Business Shows.
Oxford Road ranked the fifteen podcasts that performed best for B2B ad campaigns over the past year, using its own campaign results rather than downloads. We checked how Apple Podcasts files each show. The business genre is a minority either way, and the list says more about where buyers listen than about where an executive should be a guest.
Must Podcast Ads Be No Louder Than the Show? California Says Yes From January 2027. The Law Does Not Say What a Podcast Streaming Service Is.
California's SB 96, signed on 28 September, extends the state's loud-commercials rule from video streaming to music and podcast streaming. We read the chaptered text. It borrows the federal television standard, passed both houses without a vote against, and leaves open who in podcasting's chain of hosts, ad servers and apps is the service that must comply.
Can a Sponsor Be Swapped Into a Creator's Old YouTube Videos? YouTube Now Lets One Brand Segment Run in Twenty Videos at Once. The Disclosure Rules Assume a Video Stays the Same.
At Made on YouTube on 23 September, YouTube said it is extending dynamic brand segments: sponsorship slots that can be inserted into up to 20 long-form videos at once, targeted by region, and swapped when a deal ends. We read the announcement beside YouTube's paid-promotion rules and the FTC's Endorsement Guides. For a brand, the disclosure has to travel with the segment; for an executive interviewed by a creator, the video around their words can acquire a sponsor later.
Can an Executive Trademark Their Voice Against Deepfakes? Italy's Prime Minister Has Filed a Four-Second Clip With the EU. Trademark Law Reaches Use in Trade, Not Every Fake.
Giorgia Meloni has applied to register a recording of herself saying “I am Giorgia Meloni” as an EU sound mark, after deepfake images of her circulated. We read the EU trade mark regulation it would sit under. A registration would protect that sign in trade for the services listed. It would not, on its own, stop a cloned voice in a fraud call or a political fake.
How Did a Cloned Voice Get a Bank to Wire €95 Million? The Chief Executive Was Impersonated by Text. The Voice That Was Cloned Belonged to a Lawyer.
Reuters reported on 26 September that fraudsters took €95 million from Fideuram, Intesa Sanpaolo's private bank, with a WhatsApp message posing as the group chief executive and a call using an AI copy of a law-firm partner's voice. We read the reporting beside the FBI's warnings. The attack went for the verification step, and the FBI's advice to limit public audio runs straight into executive visibility.
Has Apple Podcasts Stopped Curating by Country? We Read Six Storefronts' Public Pages. Australia and New Zealand Shared 118 of 124 Picks.
Podnews reported on 2 October that the Apple Podcasts app now gives signed-in users an algorithmic New page rather than a country's editorial picks. We could not test the app, so we compared the public New pages for six storefronts on 5 October. They are still curated, but by region: one shelf of 20 episodes was identical in five countries, and Canada's new-show shelf was mostly America's.
Will Editors Now Verify Who Wrote Your Op-Ed? A Contributor Nobody Could Confirm Was Published Across Canada. One Paper Now Requires Writers to Meet Its Staff in Person.
Policy Options and the Ottawa Citizen withdrew articles published under the name Daniel Robson after failing to verify the writer, and changed how they accept opinion pieces. Google's guidance now calls fabricated author profiles deception. We read the editors' notes and what they mean for a company placing an executive's byline.
Will YouTube Shorts Bury Reposted Podcast Clips? YouTube Says It Will Cut the Reach of Re-uploads That Add Nothing, and Its Rules Say Permission Does Not Count.
On 1 October YouTube said Shorts recommendations will favour original content and reduce the reach of videos re-uploaded from other creators. We read the announcement and the reused-content policy behind it. For a company whose executive appeared on a podcast, the show's clip is the original; an unchanged copy on the company channel is what the change targets.
Would Listeners Accept an AI Version of a Podcast Host? Two Thirds of Weekly Listeners Chose a Negative Word. A Third Said Deceived.
Edison Research asked US weekly podcast consumers how they would feel if their favourite host used an AI-generated version of themselves. We read the results beside Apple's rule requiring AI hosts and synthetic voices to be disclosed in every episode, and YouTube's exemption for a creator cloning their own voice. For an executive podcast, the line runs between AI that helps and AI that stands in.
What Changed in the Podcast Download Rules, and Do Video Episodes Count? The New Guidelines Say Yes, Through RSS. They Also Admit the Same Log Can Give Three Different Counts.
IAB Tech Lab finalised version 2.3 of its podcast measurement guidelines on 29 September. We read the full text. Video episodes in podcast feeds now fall under the one-minute download rule, vendors may disagree by 5 per cent, three de-duplication windows are all allowed, and a definition of a play is still a 2027 job.
Can You Tell Which AI Sent You a Visitor? Gemini Has Started Tagging Its Links. Clicks From Google's AI Overviews Still Arrive Looking Like Ordinary Search.
Gemini links began carrying tracking tags at the end of September, and Google's John Mueller confirmed he could see them. We read what OpenAI, Google Analytics and Search Console each document about AI referrals. Assistants can now be separated in analytics; Google's own AI answers cannot, and a tag reported on AI Overview links turned out to come from a browser extension.
How Long Does Google Take to Show a Fix? Its Own Figures: One to Two Days for a New Title, About 30 Days to Recrawl a Known Page, Three to Six Months to Recover From a Core Update.
Gary Illyes showed Google's internal timings for 23 crawling, indexing and serving processes at Search Central Live in Barcelona on 2 October. Google has not published the slides. We worked from attendees' write-ups and set the figures against Google's own documentation, picking out what matters when coverage about a company is wrong, out of date or needs to come down.
What Did Google Change in Its Content Guidance on 1 October? We Compared the Pages. The Changelog Lists One Edit. There Were Two, and the Unlisted One Calls Invented Authors Deception.
Google's changelog records three new sentences telling sites to fact-check AI-written content by hand. We diffed both of Google's content guidance pages against June copies. Its people-first content page grew by 458 words with no changelog entry: a definition of main content, effort, originality, skill and accuracy, and a warning against fabricated creator profiles. What it means for ghostwritten bylines and contributed articles.
Is ChatGPT a Search Engine Under EU Law? The Commission Says Yes. It Lists 159.1 Million EU Users, More Than Bing, and the Heavier Rules Apply From January.
ChatGPT is the first AI assistant designated a very large online search engine under the Digital Services Act. We read the Commission's decision notice, its register of 28 designated services and the articles that now apply. Risk assessments, independent audits, researcher data access and, if ChatGPT shows ads in the EU, a public ad library. What changes for a company that appears in ChatGPT's answers, and what does not.
Where Does the New Siri Get Its Answers About a Company? Apple Says From Its Own Crawler. We Checked 30 News Sites: 22 Block Apple's AI Training, Only One Blocks Siri.
Siri AI now answers broad questions from the web on iPhone, iPad and Mac. Apple's Applebot documentation sets out three separate controls: one for search, one for training and one for AI answers. On 5 October we read the robots.txt files of 30 major news and business outlets to see which of them coverage of a company can reach Siri through. Nearly all of them, with one notable exception.
Is Netflix Now a Podcast Platform? A Fifth of Podcast Consumers Used It in Its First Year. Only 5 Per Cent Call It Their Main One, and No Show Can Put Itself There.
Sounds Profitable's Podcast Landscape 2026, out on 1 October, puts Netflix third among podcast platforms behind YouTube and Spotify and ahead of Apple Podcasts, in its first year carrying shows. We read the 27-slide deck. The bigger shifts for a company booking guest spots are YouTube's hold on the television, shrinking podcast apps and listeners who follow only two or three shows.
How Does LinkedIn Decide Who Sees an Executive's Post? Its Engineers Say Your Network Supplies Over 70 Per Cent of the Feed. Now It Is Testing a Tab for Nothing Else.
LinkedIn's chief product officer announced a test on 18 September that splits the feed into a Network tab and a For You tab. The same day, a paper by LinkedIn engineers said a member's own network supplies over 70 per cent of feed impressions and engagement. We read what LinkedIn has said about how the feed picks posts, what that means for an executive, and which popular claims LinkedIn has never made.
How Does a False Wikipedia Edit Reach Google? A Living Wrestler Was Declared Dead Seven Times on Two Wikipedias. We Timed Every Edit.
WWE's Karmen Petrovic was edited into death on Portuguese and English Wikipedia on 29 and 30 September, and Google was reported to show the same false date. We pulled every revision. English claims lasted 5 to 54 minutes, the Portuguese one five and a half hours, Wikidata was never touched, and one editor cited Google itself as the source.
Is AI Pulling Readers Away From Wikipedia? We Pulled Wikimedia's Own Pageview Data. Human Views Fell 7.5 Per Cent in a Year, and a Third of Pageviews Are Now Bots.
The Wikimedia Foundation told the UN General Assembly on 30 September that AI is straining Wikipedia while sending fewer people to it. We pulled English Wikipedia's monthly pageviews from 2019 to September 2026. September was the lowest month in the series, down 9.3 per cent on a year earlier even after Wikimedia's 2025 bot-detection fix.
Can Google AI Mode Monitor News About Your Company? Anyone Can Now Ask It to Watch a Topic. Google Has Not Said How It Picks the Sources.
On 28 September Google began rolling out AI Mode's information monitoring to all users. You describe what to watch; Search checks sites, forums and social posts and sends a written update. We read Google's own posts and set it against Google Alerts: what it could add to a company's monitoring, what it cannot yet do, and why the bigger change is other people monitoring you.
Can You Still Monitor Reddit for Mentions of Your Company? We Tested Five Logged-Out Routes. Only RSS Still Works, and It Ends on 13 November.
Reddit announced on 30 September that it will stop serving RSS feeds on 13 November and close its public API in March 2027. We tested every logged-out way a script can read Reddit today. JSON, the API and Old Reddit are already shut; RSS keyword search still returns results, for six more weeks.
What Happens to a Business Caught Buying Fake Reviews? In Singapore, Six Months of Public Apology. The Reviews Cost About S$6.30 Each.
Singapore's consumer watchdog named 45 businesses that bought AI-written reviews from one seller at S$139 for 21. Each must apologise publicly for six months. We read the annexes, priced the packages and checked the seller's own site, then set the case beside the FTC's warning letters and the UK's new fake-review powers.
Will California's Journalism Tax Credit Put More Reporters in Newsrooms? It Pays Up to $20,000 a Reporter. It Measures Dollars, Not Jobs.
Governor Newsom signed AB 2222 on 30 September. From 2027 local outlets get refundable credits of $20,000 for each of their first five full-time journalists and $15,000 after that. We read the enrolled text: no annual cap, a corrections-policy requirement, and performance indicators that count money paid out rather than reporters employed.
Can a Platform Be Held Liable When a Scam Ad Impersonates Your Company? Not Yet. The FTC Has Just Asked 21 Questions About Making It So.
On 1 October the FTC published an advance notice proposing to make search engines, social networks and marketplaces answer for impersonation scam ads their tools optimise. It is the first step of a process that last time took 26 months. What it proposes, what Section 230 has to do with it, and why a company that has been impersonated has until 30 November to say so.
Is a Stacker Pickup Earned Media? We Read 83 Brand-Funded Stories on 27 News Sites. None Was Labelled Sponsored, and Every Link to the Brand Was Followed.
Stacker sells brands distribution of their stories to 3,000 newsrooms as “earned”, and says readers do not know they are sponsored “because they're not”. Brands pay; publishers get the stories free. We read 83 live pickups: each names the producing brand in a closing line, none carries a sponsored label, all 466 links to the producing companies pass ranking credit, and 29 company stories point search engines back to the brand's own page.
How Long Does a Google Spam Update Take? We Timed Every Update on Google's Own Dashboard.
We read all 42 ranking incidents Google has posted. Spam updates since 2023 have run from 19 hours to 27 days, eight of 17 updates given a stated window ran past it, and 2026 has already had four spam updates, more than any year on the record.
Does Google Pay Publishers When AI Overviews Use Their Work? About 100 Get Paid. Everyone Else Got a Toggle.
An invite-only pilot pays roughly 100 sites, some under $1,000 over months and one over $1 million a year, by a formula nobody outside Google has seen. Everyone else has an opt-out switch, and on 30 September a federal judge dismissed the lawsuit arguing traffic was owed.
Can You Pay to Appear in a ChatGPT Answer? You Can Pay to Appear Under One.
OpenAI sells ads in 63 countries and is testing Sponsored Agents a user can talk to. Its own documentation says advertisers have “no ability to shape, rank, or alter” the answer itself. What the ad buys, what it cannot, and what the advertisers say about measuring it.
How Many Expert Pitches Are Written by AI? One Platform Says 21 Per Cent. We Checked Where a Fake Expert's Quotes Still Stand.
Featured scored 1.39 million source pitches and flagged 21 per cent as fully AI-generated. We retested the ten articles Press Gazette linked for an AI-generated “art therapist”: two weeks on, four still quote her and only one carries a correction.
Can a PR Survey Use AI Respondents? Pew Tried It on Nearly 300 Questions. The Answers Were 12 Points Off.
Pew built AI “digital twins” of real panelists and ran three of its 2026 surveys through them. The synthetic results missed by 12 points on average, turned majorities into near-unanimity, and changed depending on which model was used. That is the exact number a survey-led pitch puts in its headline.
How Does Wikipedia Decide to Delete an Article? We Read a Week of Them: 65.7 Per Cent of Company Articles Went.
We pulled all 544 deletion discussions from the first week of September. Company articles were deleted at 65.7 per cent against 43.1 per cent for everything else, the median discussion drew two responses, and ten editors closed 82.5 per cent of them.
Do Preprints and Open-Access Papers Get Cited by AI? arXiv Blocks Nobody. Nature Blocks Thirteen.
A robots.txt audit of 17 scholarly hosts against 17 named AI agents. arXiv, PubMed Central, Zenodo, PLOS and OpenAlex name none; Nature and Springer block thirteen each. No one publishes a citation rate for academic sources, and this guide does not invent one.
Where Does a Wire Release Actually End Up? 45 Per Cent of the Partner Copies Were Already Gone.
Partner copies self-canonicalise away from the wire, drop the release's body links entirely, and carry a pickup-counting beacon the wire's own page does not. Of the syndicated copies in one crawl sample, 45.1 per cent returned 404 against 0.2 per cent on the wire itself.
Do Investor Materials Get Cited by AI? The SEC Invites the Crawlers In. The IR Sites Do Not.
EDGAR's robots.txt explicitly allows the filing archive and crawlers take it, across 7,163 companies at a 0.5 per cent loss rate. Five unrelated megacaps serve the same 207-byte vendor robots.txt naming no AI crawler at all.
Do Books Build Authority? Amazon and WorldCat Block the Crawlers Site-Wide.
An ISBN is a supply-chain identifier that confers nothing, the bestseller list is three self-selected categories, and the two catalogues that would vouch for a book disallow GPTBot, ClaudeBot and the rest outright.
Do Employer Reviews Affect Brand Search? Glassdoor Blocks 21 AI Agents and Its Own Guidelines.
Glassdoor's robots.txt sorts AI crawlers into high and low quality in its own prose and blocks all of them from its reviews - and disallows its community guidelines to everyone, which is why this guide quotes no platform policy text.
What Happens to Coverage When a Journalist Leaves? The Byline Survives. The Masthead Is the Risk.
We tested 57 author pages captured in 2016. TechCrunch returned 16 of 16, Ars 17 of 18, and bylines and Person markup survived intact. The catastrophic case is not departure but closure: Protocol now redirects to a domain-sale listing.
How Do You Fix Entity Confusion Between Similar Companies? 139 Things Are Called Atlas.
Our own Wikidata queries: 266,260 items typed as a business, 143,534 with a website, and just 4,404 carrying the statement that actually disambiguates them from something else. What identifiers assert, and what no company can do but wait for.
Does a Google Business Profile Affect Brand Search? Not One of the Eleven Metrics Tells You.
Google documents eleven profile metrics and none separates branded from unbranded demand, so the literal question cannot be answered with Google's own tooling. What the profile does control, and which companies are eligible at all.
Does Non-English Coverage Affect AI Visibility? Models Answering Hindi Queries Cite English Wikipedia Most.
A Stanford evaluation across six chatbots and 2,100 questions found an Anglophone retrieval bias, and retrieval rather than reasoning drives over 70 per cent of errors. English is 41.9 per cent of the crawlable web against Hindi's 0.21.
Do Analyst Firms Affect What AI Recommends? Forrester Fences Off Its Reports and Leaves the Reprints Open.
Forrester's robots.txt names eight AI crawlers and blocks them from /report/ alone, so the crawlable copy of an analyst report is usually the vendor-paid reprint on the vendor's own domain. Gartner is Perplexity's third most-cited source and absent from ChatGPT's top ten.
Do Trade Publications Outperform National Outlets? TechRadar Out-Cites Reuters.
In the largest published citation dataset, TechRadar takes 0.9 per cent of ChatGPT citations against Reuters at 0.6, and no general-interest newspaper of record appears in any platform's top ten. Reach and relevance are not the same unit, and no dataset converts between them.
Do AI Agents Fill Out Your Contact Form? Nobody Has Published a Trustworthy Number.
Agents that click, navigate and submit are arriving, but the volume is unmeasured: the firms counting it sell the defence against it, and an agent running as an extension in someone's own Chrome has no user agent, no IP range and no signature to count.
How Long Does Google Take to Index a New Page? Google Only Answers a Different Question.
The one figure Google publishes - a few days to a few weeks - is about crawling, not indexing, and it twice refuses to guarantee indexing at all. The four states a URL passes through, why a page is crawled and never indexed, and what is documented to help.
Do Conference Talks Build Authority? The Schema Has No Speaker Property.
Schema.org's Event type defines no speaker, only performer - which Google glosses as artists and comedians - while it does define recordedIn. The durable value is the artefact rather than the room, and no study measures whether speaking moves any observable outcome.
Do Reddit Threads Show Up in AI Answers? Reddit Leads the Citations, at 2.2%.
Reddit is the single most-cited domain in Google AI Overviews and Perplexity on the largest published citation dataset, and Google and OpenAI both pay for structured access. The share is far smaller than the headlines suggest, and every measurement of it comes from a vendor.
Do Review Sites Affect What AI Recommends? One Company Now Owns Four of Them.
G2, Gartner, Yelp and TripAdvisor each sit at around 1% of assistant citations. G2 now owns Capterra, Software Advice and GetApp and licenses its reviews to OpenAI directly. Nobody has published a test showing a review moves a recommendation.
Does a YouTube Presence Affect AI Answers? Pew Found Video Is Cited Less, Not More.
YouTube links are somewhat more common in ordinary Google results than in AI summaries. No video platform appears in the ten most-cited domains across six chatbots, and Google's video documentation never mentions transcripts.
How Many Pitches Does a Journalist Get? The Surveys Say Five a Day. The Logs Say 3.69% Get a Reply.
Journalists say they prefer pitches before noon; the email logs show replies peak between 6pm and midnight. Every survey in this field is published by a company selling pitching software to the people it surveys.
Do Press Embargoes Still Hold? Only Where Someone Signed Something.
An embargo binds a journalist only where they signed for the material. The documented record shows breaks are handled by lifting the embargo for everyone, and that sanctions on the outlet that broke it are rare.
Do AI Overviews Reduce Clicks? Pew Says the Rate Halves. Google Says the Study Is Flawed.
Users clicked a result on 8% of Google searches carrying an AI summary against 15% without, and clicked a link inside the summary on 1% of visits. Google publishes no data of its own.
How Often Do AI Search Tools Cite the Wrong Source? More Than 60% of the Time.
The Tow Center ran 1,600 excerpt queries through eight tools. Perplexity was wrong 37% of the time, Grok 3 was wrong 94% of the time, and almost none of them admitted uncertainty.
How Much Traffic Do AI Assistants Actually Refer? Anthropic Crawled 38,066 Pages Per Visit.
Cloudflare's crawl-to-refer ratios put OpenAI at 1,091:1 and Google at 5:1. On the returning side, AI platforms sent 35.9 million visits to news and media sites in a month when search referrals were falling by hundreds of millions a year.
What Do People Actually Ask AI Chatbots? Three Topics Cover Roughly 77% of Messages.
A 1.1-million-conversation study published through NBER with OpenAI authors puts Seeking Information at 24% of ChatGPT messages and rising. The share about anything purchasable is never printed.
Do People Trust What AI Says About Companies? Twenty Per Cent Say Yes, and Four Per Cent Click the Source.
The Reuters Institute surveyed 97,520 people across 48 markets. Trust in AI chatbot answers is 20%, against 37% for news overall, and almost nobody clicks through to check it.
Can You Tell an AI Not to Train on Your Site? The IETF Is Standardising How to Ask.
A working group is building the vocabulary for refusing AI training. Its charter puts enforcement explicitly out of scope, and the specification says a recipient has a choice whether to follow what you express.
Platforms Filed Nearly Four Billion Moderation Notices in Six Months. Researchers Found Most Are Templates.
The EU's transparency database holds 3,944,217,973 statements of reasons from 373 platforms, 43 per cent of them fully automated. The volume is real. The explanatory value is contested by the academics who read them.
What Does NewsGuard Actually Measure? Twenty-Five of Its Hundred Points Have Nothing to Do With Accuracy.
Nine criteria, weighted from 22 points down to 5. Four cover whether a site tells the truth. Four cover whether it tells you who owns it, who writes it and what is an advertisement.
Can You Charge an AI Crawler? Cloudflare Built a Toll Booth on a Status Code Nobody Used.
Pay Per Crawl answers a request without payment using HTTP 402, a response code reserved in 1997 and left unimplemented ever since. It is in closed beta, and Cloudflare's documentation names no AI company paying.
Google Now Reports AI Impressions. It Does Not Report the Clicks.
The generative AI performance report shows how often your links appeared in AI Overviews and AI Mode. It shows no clicks, no click-through rate and no queries — and the impressions were already inside your existing totals.
What Is Expired Domain Abuse, and Does Buying an Aged Domain Trigger It?
Google's spam policies name expired domain abuse as a distinct violation. The definition turns on purpose, not on the age of the domain — and the policy says outright that reusing an old domain for a genuine new site is fine.
Do Newsletters and Substack Count as Media Evidence for a Visa Petition?
The regulation names no platform at all. It sets a test — circulation, readership, intended audience, and whether the piece is genuinely about you — and a newsletter passes or fails on the same terms as a newspaper.
Which Publishers Have AI Licensing Deals, and What Are They Worth?
We counted sixteen publisher–AI deals announced across 2025 in one trade timeline. Financial terms were disclosed for none of them. The few figures in circulation are reported estimates, not confirmed sums.
What Are Content Credentials, and Has Any Newsroom Actually Deployed Them?
C2PA's own deployment guidance reports 54 conforming products from 31 companies. It names a US Defense Department platform, LinkedIn and YouTube — and not a single news publisher.
Does the EU AI Act Require You to Label AI-Written Coverage?
Article 50 applies from 2 August 2026 and requires AI-generated text on matters of public interest to be disclosed — with an exemption that turns on human editorial control and named editorial responsibility.
Do Wire Service Links Pass SEO Value? We Checked 70 Releases.
Across 70 PR Newswire releases, only eight contained a single external link in the release body at all. Every one of those nineteen links carried rel=nofollow. Not one passed ranking credit.
Does Anyone Actually Publish llms.txt?
We probed 41 domains. Not one news publisher serves an llms.txt file. Developer tooling companies serve them almost universally. And the proposal itself never claims any AI system reads them.
Which AI Crawlers Do Publishers Actually Name?
We read the robots.txt of 25 publishers and counted every AI agent named, separating them from SEO tools. Thirty-five distinct AI crawlers appear. CNN names 29 of them. All four wire services name none.
What Sources Do Wikipedia Company Articles Actually Cite?
We extracted every citation from 24 company articles: 2,465 links across 548 domains. TechCrunch is the single biggest source by a wide margin, and Wikipedia rates it no consensus. Two in five rated citations go to sources Wikipedia does not fully trust.
What Happens to a Publisher's Credibility When It Uses AI?
We scanned all 497 entries on Wikipedia's perennial sources list for AI as a stated reliability factor. Twelve name it. The pattern is not that AI use is disqualifying, but that undisclosed AI use is, and it can take an entire publisher group down with it.
How Does an Entity Actually Get Into Google's Knowledge Graph?
Google launched the Knowledge Graph in 2012, documents what a panel shows, and lets some subjects claim one. It has never documented a route in, and the one entry mechanism it does publish carries an explicit no-guarantee clause.
Does Gemini Decide What Goes Into Google's Knowledge Graph?
No. The Knowledge Graph shipped in May 2012, more than a decade before Gemini existed, and Google's Gemini grounding documentation never mentions it. They are separate systems doing different jobs, and conflating them produces advice that cannot work.
Do Identifier Registries Like Wikidata, ISNI and LEI Affect Search Visibility?
None of the five registries we checked claims a search benefit anywhere in its own documentation. Google, however, accepts four registry identifiers by name in its Organization markup, and says some are used behind the scenes to tell your company apart from others.
Is Wikidata Easier to Get Into Than Wikipedia?
Substantially, yes. Wikipedia asks for significant coverage in independent reliable sources. Wikidata asks for a sitelink, a clearly identifiable entity with serious references, or a structural need, and its own guidance says the wording is deliberately loose.
What Are Wikipedia's Perennial Sources? We Counted All 497.
We parsed the full list on 18 September 2026. Of 497 sources Wikipedia has argued about repeatedly, 156 are rated generally reliable. More are rated generally unreliable than reliable, and 66 are deprecated outright.
How Does Wikipedia Rate Forbes? Three Entries, Three Answers.
Forbes appears on Wikipedia's perennial sources list three separate times. Print and pre-2010 Forbes.com is generally reliable. Post-2010 Forbes.com has no consensus. Non-staff contributor articles are generally unreliable and treated as self-published.
Are Tier-One Media Features Still Worth It?
We cross-referenced fifteen major outlets against two independent measures: how Wikipedia rates them as sources, and how many AI crawlers they block. Thirteen are generally reliable. Four block every AI crawler we tested.
What Actually Makes an Outlet Tier One?
Tier one is a sales category, not a standard. There is no register, no criteria and no authority. Three public measures exist that can be checked instead, and they disagree with each other often enough to be worth checking.
You Were a Podcast Guest. Why Doesn't Google Show the Interview?
We measured twenty real episode pages. They range from 276 words to 53,552 — and the thin ones carry no transcript, no schema, and sometimes sit behind a blanket crawler block.
Spotify or YouTube: Which Platform Actually Gives a Podcast Guest Search Visibility?
We measured what each platform publishes as readable text. Apple returned about 6,000 words per show page, YouTube around 2,000 — and Spotify returned four.
Why Podcast Transcripts May Matter More to AI Than the Audio
We checked 52 real feeds. Ten declare machine-readable transcripts — and 1,129 of those declarations sit on a domain that forbids every crawler.
How Many Websites Does One Podcast Episode Actually Appear On?
Spotify names eight destinations and says you must submit to each yourself. We measured what those listings contain — and the range runs from 7,258 words to four.
Can ChatGPT Find You From a Podcast Interview? Here Is the Experiment.
We will not claim we tested this, because we have not. This is the pre-registered protocol — fixed prompts, control subjects, measurement windows — so the result cannot be steered after the fact.
Putting One Executive on Ten Podcasts: The Study Design
The flagship experiment, published as a method before it is run. What would have to be measured, what would count as a result, and why a null finding is the likeliest outcome.
Does Saying Your Name and Company on a Podcast Help AI Understand Who You Are?
A testable question with a documented mechanism behind it — and a measurement we ran showing almost nobody uses the mechanism. Here is the design, and what we already know.
What Happens to a Podcast Interview After Twelve Months?
Two findings already published in this series bound the answer: attention collapses within days, and pages disappear within years. Here is the design that would measure what survives in between.
Does Appearing on a Bigger Podcast Domain Make AI Trust You More?
The obvious test is identical information on a large platform versus a small one. The obvious test is also confounded — and Google's own documentation explains why.
Spotify or YouTube for a Guest: The Test, and What We Already Know
One half of this question is already answered by measurement. The other half needs an experiment — and the measured half may make it unnecessary.
Ten Podcast Interviews or One Forbes Article: How Would You Actually Tell?
Not which is better as an assumption. The design that would measure it — and the reason the honest comparison is harder than either side admits.
Does Paid Stock Promotion Have to Be Disclosed?
Federal securities law requires it, and requires the amount. Section 17(b) is the strictest disclosure rule any placement touches — and the only one naming a dollar figure.
Can You Pay Someone to Edit Your Wikipedia Article?
Yes, if they disclose employer, client and beneficiary. Undisclosed paid editing breaches the Terms of Use, and paid editors are strongly discouraged from touching the article directly at all.
Do FAQ Rich Results Still Exist?
No. Google retired them on 7 May 2026 and deleted the documentation five weeks later. The markup is still valid schema — it just no longer buys anything in Search.
Does CAN-SPAM Apply to PR Pitches and Press Release Email?
The FTC says the law "makes no exception for business-to-business email." Penalties run to $53,088 per message, and the rules cover subject lines, addresses and opt-outs.
What Does Apple Actually Require of a Podcast?
A public RSS feed to spec, a permanent GUID per episode, ASCII URLs and at least one episode. The requirements are technical — and they decide whether a guest appearance is findable at all.
Who Is Responsible When a Disclosure Fails?
Not the platform. The FTC says responsibility "rests with the influencer and the brand" — and a built-in disclosure tool is "no guarantee" of anything.
Does Article Markup Help Coverage Rank?
Google says there is "no markup requirement" to be eligible for Top stories. Article schema tells Google what a page is — it does not buy a place in anything.
Does Blocking a Crawler Remove You From Search?
No. Google says robots.txt is "not a mechanism for keeping a web page out of Google" — a blocked URL can still appear in results, without the content behind it.
Is There an AI Exemption From Advertising Law?
The FTC's answer is four words: "there is no AI exemption from the laws on the books." Operation AI Comply has the settlements to show it means them.
How Do You Verify a Crawler Is What It Claims to Be?
By reverse DNS, not by the user-agent string. Anyone can claim to be Googlebot — and not every documented crawler follows robots.txt the same way.
Does Google's Site Reputation Abuse Policy Apply to Sponsored Articles?
It targets third-party content published on a host site mainly for that host's ranking signals. Whether your placement qualifies turns on one word: integration.
What Are Journalism's Trust Indicators, and Where Does Sponsored Content Sit?
Eight indicators, and one of them exists specifically to tell readers whether what they are reading was paid for. Placement is addressed by name.
What Happens to Coverage in a Google Core Update?
Google says core updates “don't target specific sites” and that a drop is not a penalty. It also says recovery can take months, and that there may be nothing to fix.
How Do Fact-Checkers Work, and Would Your Coverage Survive One?
The IFCN code asks signatories to identify the source of all significant evidence and to prefer primary sources. Most promotional coverage has neither.
Does Paywalled Coverage Still Get Indexed?
Yes, if the publisher marks it up correctly. The markup exists so Google can tell paywalled content apart from cloaking — which violates its spam policies.
Does Google Penalise AI-Generated Coverage?
Not for being AI-generated. The violation is automation used “with the primary purpose of manipulating ranking” — which is about why content exists, not how it was typed.
What Does Google News Require of Sponsored Content?
Disclosure, in terms. Sponsorship — including ownership, payment or material support — must be clear to readers, and content concealing it is not allowed.
Does Google Discover Carry Sponsored Content?
Yes, disclosed. Discover's sponsorship rule is almost word for word the one in Google News — the same standard applied twice, which tells you it is settled policy rather than a quirk.
What Is Scaled Content Abuse, and Does Bulk Placement Trigger It?
Google's policy targets producing many pages “primarily for manipulating search rankings and not helping users” — regardless of how they were made.
How Do Publications Handle Corrections, and Can You Request One?
Yes, and it is the request most likely to succeed. Journalism's codes ask for errors to be corrected quickly and prominently — which is a far lower bar than removal.
How Many Journalists Are There, and Is the Number Falling?
About 47,100 in the United States, and the government projects a 6% decline by 2035 — while still expecting 3,400 openings a year, every one of them a replacement.
What Does an ‘Impression’ Actually Mean, and Who Decides?
Half the pixels, for one continuous second. The standard says plainly that this establishes an “opportunity to see” — not that a human being saw anything.
Do News Sites Block AI Crawlers? We Checked Fifteen.
Twelve of fourteen block Common Crawl. Only four block OpenAI's search agents, and several explicitly welcome them. Publishers are barring training while admitting search.
What Do Business Awards Cost, and Do They Mean Anything?
Published entry fees run from $275 to $2,400 per submission, rising at each deadline. The programmes disclose this openly — the question is what a paid entry signals to anyone reading the badge later.
What Happens to Coverage When a Publication Shuts Down?
It usually disappears. MTV News took two decades of reporting offline, one weekly took 22,000 stories with it — and publishers are now closing the archive that used to catch the fall.
Will a Wire Service Publish Anything You Pay For?
No. The guidelines demand a newsworthy angle, an objective tone, documentation for legal claims, and no more than one link per hundred words — and ban a dozen categories outright.
What Happened to HARO, and Does Source-Request Journalism Still Work?
Cision closed it, then sold it. The buyer's pitch is that AI made human sources scarce — which is a good line, and comes from the party with something to sell.
What Is Common Crawl, and Why Is Everyone Blocking It?
A 501(c)(3) non-profit giving away a copy of the web for research. It is now the single most blocked crawler on major news sites — more than OpenAI's, more than Anthropic's.
What Does a Podcast Guest Appearance Cost, and What Sets the Price?
Normally nothing — guests are booked because a show wants them. Where a fee exists, reported figures run from a few hundred dollars to five figures, and every source quoting them sells bookings.
Does an O-1 Petition Treat Media Coverage Differently From EB-1A?
In one respect, strikingly. The O-1B arts criterion names advertisements and publicity releases among acceptable evidence — categories excluded almost everywhere else.
What Happens to a Placement's Traffic After the First Week?
Research on 606 news articles found traffic to about 80% of them falling continuously within twelve hours. A separate study put the median article's half-life at 36 hours.
Do Podcast Audiences Convert, or Only Recall?
Nobody can say. One benchmark found promo codes capture about 21% of conversions — and every firm publishing a figure sells the attribution product that fixes it.
How Do AI Assistants Decide Which Companies to Recommend?
Being recommended is not the same as being cited, and it is the less documented of the two. No operator publishes how a company gets named as an option.
Can a Company Correct What an AI Says About It?
Google lets a verified representative suggest changes to a knowledge panel. No AI assistant publishes any comparable way to fix a false statement about your business.
Is Sponsored Coverage Legally Different From Earned Coverage?
Yes. The FTC treats a format that looks like journalism as potentially deceptive in itself — and has alleged deception even where the page carried an “Advertorial” label.
Does Being a Podcast Guest Do Anything for Search Visibility?
An unpaid appearance produces ordinary editorial links. A paid one produces links Google's policy says must be marked so they pass no ranking credit at all.
How Quickly Does New Coverage Appear in AI Answers?
There are two clocks, and only one of them can move fast. No operator publishes either — and every timing figure in circulation comes from a firm selling AI visibility.
What Is Topical Authority, and Is It Measurable?
Google names seventeen ranking systems. Topical authority is not one of them. The idea behind it is sound; the metric sold against it does not exist.
Is There a Difference Between Ranking in Google and Being Cited by an AI?
For Google's own AI features, barely any. A page qualifies by being indexed and snippet-eligible, and Google states there are no special optimizations for it.
Can a Placement Be Removed, and Who Controls That?
The publisher does. Journalism's own guidance says seldom remove archived material — and warns that weighing a subject's influence puts a newsroom's credibility at risk.
What Is a Contributor Network, and How Is It Different From a Staff Byline?
You write it, you pay for membership, and it appears under a masthead readers trust. Forbes describes the benefits in detail and publishes no price at all.
Do Author Bylines Affect Credibility Signals?
Google strongly encourages accurate bylines — and says in the same documentation that E-E-A-T is not a ranking factor and rater data never touches the algorithm.
What Share of Podcast Listeners Finish an Episode?
Apple counts an engaged listener at 20 minutes or 40%. The IAB counts a download at roughly 60 seconds. The two industries are measuring different things and reporting one number.
Does Coverage in a Major Outlet Change How Customers Search for a Brand?
Probably, a bit, briefly. But the 40% and 50% lift figures in circulation cannot have come from the tool everyone cites, because that tool reports no search volumes at all.
What Signals Do Answer Engines Use to Judge a Source?
Microsoft now shows you which of your pages Copilot cited — and warns the number means nothing about authority. No operator explains how sources are actually chosen.
Does Media Coverage Affect B2B Buying Decisions?
The study everyone cites was produced by a PR firm and a social platform, measures content companies publish themselves, and is not about press coverage at all.
Does Podcast Audio Get Indexed and Quoted by AI Search?
The audio is the least durable part of a podcast appearance. Google built episode transcription, then shut the product down — and the schema for episodes still has no transcript field.
Does Media Coverage Get a Company a Wikipedia Article?
Wikipedia's guideline for companies excludes press releases, sponsored articles and anything substantially derived from them. Its operative sentence is four words long.
How Big Is a Podcast Audience, and Why Are Download Numbers Misleading?
The industry's own measurement guideline defines a download as roughly one minute of audio having been requested. It does not, and cannot, mean anyone listened.
Do the Wire Services Let AI Crawlers Read Press Releases?
We checked what the major wire services actually tell AI crawlers. Four of five permit them by default, one publishes a file inviting them, and none of that proves a model will cite you.
Does a Google Knowledge Panel Require Media Coverage?
Google documents what a knowledge panel is and how to claim one. It does not document how to make one appear — and that silence is the whole answer.
How Do Podcast Charts Work, and Can a Ranking Be Bought?
Apple names three signals, excludes three others, and declines to publish the weighting. Almost everything else you have read about podcast charts is invention.
What Is Domain Authority, and Should It Guide Outlet Selection?
Google's own guide names seventeen ranking systems. None of them is domain authority, which is a third-party prediction sold as a property of your target.
How Must Paid Content Be Labelled, and Which Labels Does the FTC Reject?
“Promoted” is not good enough. The FTC names the words that work, the words that do not, and where on the page the label has to sit.
What Is the Difference Between a Wire Release, a Contributed Post and Earned Coverage?
Three products are sold under one word. They differ in who writes them, who pays, and how every system downstream treats them — which is the part that decides what they are worth.
How Long Does a Placement Stay Findable?
Pew checked a million webpages. More than a third of what existed in 2013 was gone within a decade, and the most-trafficked news sites decay at the same rate as the least.
Does Schema Markup Affect Whether a Page Is Cited?
Google is unusually blunt about this. Structured data makes a page eligible for a feature. It does not guarantee the feature, and losing it does not change how the page ranks.
Do Sponsored Articles Count as Media Evidence for a Visa Petition?
USCIS excludes marketing material the person or their employer paid for, even when it reads as objective. The regulation, the policy manual, and what that leaves.
Does a Paid Placement Pass SEO Value, or Does Google Discount It?
Google's spam policy names advertorials directly. A correctly disclosed paid link passes no ranking credit by design, and an undisclosed one is a policy violation.
Do Readers Trust Journalists, and Has That Number Moved?
Trust in national news organisations fell 11 points in six months to 56%. Local news holds at 70%. Two large surveys disagree about the level and agree about the direction.
Video or Audio: How Are People Consuming Podcasts Now?
57% of Americans have both listened to and watched podcasts, and roughly 71% of US creators make video versions. The advertising figures still count the medium as audio.
How Do Editors Decide What to Cover, and What Do Journalists Want From a Pitch?
Seven in ten journalists rank fit with their existing coverage first. Fewer than half want original data. Almost seven in ten want the pitch under 200 words.
Do Host-Read Podcast Ads Work Better Than Produced Spots?
Nielsen reports a 50% lift in purchase and recommendation intent. The recall figures quoted everywhere else do not appear in anything Nielsen has published.
Which Sources Do AI Assistants Actually Cite?
A Stanford evaluation of six chatbots found English Wikipedia the most-cited domain globally, at 3.5% of citations. The vendor figures circulating are far larger and unsourced.
Do Press Releases Still Get Read, and by Whom?
Two thirds of journalists use PR-provided material for story ideas. Most receive 50 or more pitches a week and reject 86% of them for irrelevance.
How Should a Company Measure Whether Placement Worked?
The industry's own standards body has said for fifteen years that advertising value equivalency is invalid. It remains the number most buyers are shown.
What Does a Press Placement Actually Cost Across Outlet Tiers?
A 1,678-outlet rate card runs from $150 to $167,900. Half of it sits between $1,000 and $2,500, and the average is 45% above the typical price.
How Many Americans Actually Listen to Podcasts, and Is the Audience Still Growing?
Fifty-eight percent of Americans aged 12 and over listened to a podcast in the past month, a record. The growth is real, and the measurement has limits worth knowing.
Two IAB Studies Disagree About Podcast Ad Growth. Which One Is Right?
One IAB publication reports podcast advertising grew 17.6% in 2025. Another puts the same year at 7.9%. Both are from the IAB, and both are defensible.
What Does the FTC Require When a Placement Is Paid For?
Paid coverage is lawful. Undisclosed paid coverage is the problem the rules address, and since 2023 the liability reaches the firms that arrange it.
AI Search & Visibility
How AI assistants build their answers, and how to get into them.
AI Answer Visibility: What ChatGPT and Gemini Say About You
When someone asks an AI assistant who you are, it answers in seconds, from sources you may never have seen.
How to Get Your Company Mentioned by ChatGPT
No submission form, no paid listing. The five signals that put companies into AI answers.
Why Your Company Doesn't Appear in ChatGPT
The seven fixable gaps behind almost every missing company, in order of likelihood.
How Does ChatGPT Decide Which Companies to Recommend?
The observable pattern behind AI recommendations: entity clarity, citations, consistency, depth.
Does ChatGPT Read Press Releases?
Sometimes. Why syndication is not authority, and what actually moves the answer.
Do Podcasts Help You Rank in AI Search?
Transcripts, episode pages, and guest bios: the machine-readable trail every interview leaves.
SEO vs AEO vs GEO: What Businesses Actually Need
Three acronyms, one record. Where traditional SEO ends and AI visibility begins.
Search and AI Monitoring: Watching What the Machines Say
Rankings are half the picture. Track what the answers actually say about you.
Schema and Structured Data: Telling Search Engines Who You Are
The invisible markup that labels your content for machines, and why most sites skip it.
Topical Authority: Owning a Subject, Not a Keyword
Why repeated association with one subject beats scattered mentions everywhere.
SaaS & Cybersecurity
Authority strategy for the categories where AI research already decides shortlists.
AI Search Visibility for Cybersecurity Companies
Security buyers shortlist with AI before analysts hear about the deal. How to be in the answer.
Why SaaS Companies Need Third-Party Authority
Your website is the least trusted source about your company. Coverage is the currency.
SEO vs AI Search for SaaS Companies
What survives from classic SaaS SEO, what changes, and where the budget shifts.
How AI Search Changes B2B Software Discovery
The dark funnel gets darker: shortlists now form in one AI answer.
How Founders Can Build an Online Authority Footprint
The founder is the company's first entity. The build order, step by step.
Reputation & Search
What Google shows when someone looks you up, and how to shape it.
Owning Page One When Someone Googles You
The first page of your name is your real homepage, whether you built it or not.
The Google Knowledge Panel: The Box Beside Your Name
The information box beside known names, and the signals that earn it.
Wikipedia and Reference Presence: The Sources AI Trusts Most
Reference sites carry outsized weight with search engines and AI models alike.
Reputation Signals: What Machines Count as Credibility
The measurable signals machines treat as credibility, and how they compound.
Executive LinkedIn Presence
Usually the first personal result, and a heavy source for machines. Treat it as a publication.
Digital PR
Buying and using press coverage without getting played.
What Does It Actually Cost to Get Featured in a Major Publication?
Almost nobody publishes a price. We publish 1,682 of them.
Do Paid Press Placements Hurt Your SEO?
Google's position on paid links is public and unambiguous. What it means in practice.
Forbes Councils vs Earned Forbes Coverage: What You Are Actually Buying
Both put your name on Forbes.com. They are not the same asset.
Author and Expert Bylines: Named Credibility That Compounds
A quote is good. Your name in the author field is better.
Podcast Authority
Interviews as assets: booking, footprint, and repurposing.
Podcast Footprint: The Interview Trail That Compounds
Every episode leaves transcripts, pages, and bios that machines read.
Your Podcast Interview Is Ad Creative. Most People Let It Die on the Feed.
Advertisers fake the podcast look because it converts. You own the real thing.
Content Repurposing: One Interview, Weeks of Content
One hour of talking, cut into a month of material.
Immigration Media
Media evidence for EB-1A and O-1 petitions, built and documented. Not legal advice.
EB-1A Media Evidence: What Actually Counts as Published Material
Criterion 3 asks for published material about the person. Much impressive-looking press fails it.
What Counts as Published Material for EB-1A?
About the person, in qualifying media, with title, date, and author. Each phrase does work.
What Strong EB-1A Media Coverage Looks Like
Strong versus weak evidence, pattern by pattern.
Media Coverage for O-1 Visa Petitions
The O-1's own published material criterion, and how it differs from EB-1A.
Do Sponsored Articles Count Toward EB-1A Evidence?
Disclosure, editorial control, and subject focus decide. The honest spectrum.