Tier one is a sales term
There is no standards body for media tiers. No register, no accreditation, no published criteria, no appeals process. The term arose in agency practice as shorthand for outlets a client's board would recognise, and it has never meant anything more precise than that.
Its practical function is pricing. Tier one justifies a higher figure, and because nobody defines it, the boundary is set by whoever is selling. The same publication routinely appears as tier one on one rate card and tier two on another, with no contradiction to resolve because there was no definition to breach.
This is not necessarily dishonest. Recognition is a real property with real commercial value, and a buyer who wants a logo their customers recognise is buying something genuine. The difficulty is that recognition gets used as a proxy for credibility, reach and machine visibility, which are separate properties that do not reliably travel with it.
If the label cannot be verified, the sensible move is to replace it with measures that can.
Three measures you can actually check
The first is source status. Wikipedia's perennial sources list records whether a large adversarial community considers a publication reliable, and it is free to consult. When we parsed it on 18 September 2026 it classified 497 outlets, of which under a third were rated generally reliable.
Number of entries by status on Wikipedia:Reliable sources/Perennial sources, read October 10, 2026
Source: Wikipedia, Reliable sources/Perennial sources, table rows parsed October 10, 2026.
492 of the 493 table rows; one entry carries a separate status. Blacklisted sources are counted under their underlying status. The September 18 parse counted 497 status markers; the list changes as editors add and merge entries.
The second is machine access. An outlet's robots.txt states which AI crawlers it permits, and takes one request to read. Our measurement of fifteen major domains found everything from four outlets blocking all six agents tested to two blocking none.
The third is price, which is the weakest of the three because it is the one the seller sets. Our analysis of a 1,678-outlet rate card found placements from $150 to $167,900, with about half between $1,000 and $2,500, a distribution reflecting what the market will bear rather than any property of the journalism.
None of the three measures editorial quality directly, and none should be treated as a complete account. They are simply checkable, which the tier label is not.
What none of the three measures captures is editorial quality directly. No public metric exists for how carefully a given piece was reported, and the measures here are proxies that correlate imperfectly with it. Wikipedia's status comes closest, because it is the product of editors arguing specifically about whether a source can be trusted, with reasons recorded. It remains a summary of discussions rather than an audit of newsrooms.
The measures disagree with each other
If tier one described a coherent underlying property, these measures would broadly agree. They do not. Outlets rated generally reliable by Wikipedia include both the publications that block every AI crawler we tested and the publications that block none.
Two of the most heavily marketed tier-one destinations, Forbes and Business Insider, are the two in our set that Wikipedia rates at no consensus rather than generally reliable. Attainability and source standing are related, and not in the buyer's favour.
Price tracks neither measure reliably. Nothing in a rate card tells you whether an outlet's contributed section is classified as self-published, or whether its robots.txt bars the agents you were hoping would read the piece.
So a single tier label is bundling at least three properties that move independently. That is precisely the condition under which a label stops being informative and starts doing rhetorical work.
The masthead is often the wrong unit
The deeper problem with tiering publications is that publications are not uniform, and the most demanding public assessor of sources does not treat them as such. Wikipedia splits twenty-one outlet families into multiple entries.
Sometimes the split is by byline. Forbes has three entries, with non-staff articles — including Forbes Councils — rated generally unreliable and treated as self-published, while print and pre-2010 Forbes.com is generally reliable.
Sometimes it is by section. Bloomberg News is generally reliable while Bloomberg's company and executive profiles sit at no consensus, with explicit consensus that they “should not be used to establish notability”. The Guardian's blogs are rated apart from its reporting.
Sometimes it is by era. CNET is split three ways by ownership, and Newsweek at 2013. An outlet's tier is treated as a fixed attribute, but the assessments that matter are conditional on when the piece ran, which section carried it and who wrote it.
Entrepreneur's entry shows how these judgements compound. It records no consensus for the magazine overall, consensus that “contributor pieces in the publication should be treated as self-published”, and editor concern that its coverage “tends toward churnalism and may include improperly disclosed paid pieces”. A single tier label for that publication would have to average a reasonable magazine, a self-published channel and a suspected undisclosed-advertising problem into one number.
Crawler policy splits below the masthead too, and along revealing lines. The Washington Post's robots.txt bars AI agents from its journalism while carving out exceptions for its /creativegroup/ and /advertising/ paths — machines are refused the reporting and welcomed to the commercial pages. Bloomberg runs a similar pattern, blocking broadly while allowing several corporate and professional paths. Tiering a domain misses this entirely, because the relevant unit is the path.
A better question than tier
Replace “is this outlet tier one” with four specific questions. Will this be staff-written or contributed? Which section will carry it? How does Wikipedia rate that specific category of content? And does the domain permit the crawlers whose systems I care about?
Those four are answerable before money changes hands, and they determine most of what the placement can actually do. A contributed piece in a famous publication and a staff feature in a specialist trade title are very different assets, and the tier label ranks them the wrong way round for several purposes.
It also helps to name the job. Human credibility, encyclopaedic standing and AI retrieval are three distinct outcomes served by different properties, and no single placement reliably maximises all three.
Where an outlet genuinely is recognised, well regarded as a source, and open to the crawlers you care about, it is worth what it costs. That combination exists. It is simply not what the word tier is telling you.
There is one narrow exemption worth knowing, because it is the only documented route by which contributed material recovers standing. Wikipedia allows that self-published articles “may occasionally be considered reliable if they qualify for the subject-matter expert exemption”. That turns entirely on the author's independently established expertise — publications, credentials, prior citation — and not on the outlet. It is earned before the article is written, and cannot be bought with the placement.
What this means for anyone buying coverage
Treat any tier described on a rate card as a pricing band, because that is what it is. Ask what the seller's tier definition is; the answer, or its absence, is informative.
Do the two free checks before committing. The perennial sources status of the specific content type, and the domain's robots.txt. Neither takes long, and together they catch the most common mismatches between what is sold and what is delivered.
Judge the outcome against the job you named in advance, rather than against the tier you were sold. If the objective was encyclopaedic standing and the deliverable was a contributed article, the placement has not failed — it was never capable of that outcome, and the mismatch happened at the briefing stage.
Media Morpheus sells media and podcast placement and publishes its own rate data. We use tier language ourselves because buyers expect it, and this report is an argument for holding that language — ours included — to measures a buyer can verify without asking us.
Methodology and limitations
This report draws on three sources of evidence, two of them our own prior measurement. Source status comes from our parse of Wikipedia:Reliable sources/Perennial sources on 18 September 2026, which classified 497 entries on the live list. Crawler posture comes from our robots.txt measurement across fifteen news domains on 19 September 2026. Price distribution comes from our published analysis of a 1,678-outlet rate card. The central claim — that no standards body defines tier one — rests on the absence of any such definition rather than on a source denying one exists; we are not aware of any accreditation scheme, and none is cited by the vendors who use the term. Limitations: robots.txt records stated permission rather than behaviour; the perennial sources list is not a policy and records consensus at a point in time; and the rate card reflects one market segment rather than all media pricing. The grouping of list rows into outlet families involves judgement at a small number of edges.
Conclusion
Tier one has no definition, no register and no authority behind it, which is why the same outlet moves between tiers depending on who is selling. Three public measures can be checked instead, and they disagree often enough to make the single label misleading: source standing, machine access and price track different things. More fundamentally, the masthead is frequently the wrong unit of analysis, because the assessments that matter attach to a section, a byline type or an ownership era rather than to a publication. A buyer who names the job, checks the specific content category and reads the robots.txt will make better decisions than one who buys a tier.
Frequently Asked Questions
Who decides what counts as a tier-one outlet?
Nobody, in any formal sense. There is no standards body, register or accreditation for media tiers. The term is agency and vendor shorthand for recognisability, and the boundary is set by whoever is pricing the placement.
What can I check instead of a tier label?
Three things that are public: the outlet's status on Wikipedia's perennial sources list for the specific content type, the domain's robots.txt for which AI crawlers it permits, and the actual price against comparable placements.
Why does Wikipedia rate parts of one publication differently?
Because editorial conditions differ within a publication. Splits occur by byline type (staff versus contributed), by section (Bloomberg's profiles, The Guardian's blogs) and by ownership era (CNET, Newsweek). Twenty-one outlet families carry more than one entry.
Is a specialist trade title worse than a famous outlet?
Not necessarily, and sometimes the reverse. A staff-written feature in a generally reliable trade publication can outperform a contributed article in a household-name outlet for both encyclopaedic standing and machine access.
Does a higher price mean a better placement?
Price is the weakest of the three measures, because it is the only one the seller controls. Our rate card analysis found placements from $150 to $167,900 with about half between $1,000 and $2,500, a spread that reflects market positioning rather than journalism.
Sources
- official Wikipedia. Reliable sources/Perennial sources: status classifications, and the separate entries for Forbes, Bloomberg profiles, The Guardian blogs, CNET and Newsweek.
- vendor research Media Morpheus, September 2026. Original parse of the perennial sources list on 18 September 2026 and original robots.txt measurement across fifteen news domains on 19 September 2026. Methods and limitations stated above.
- vendor research Media Morpheus, 2026. Analysis of a 1,678-outlet placement rate card, source of the $150 to $167,900 range and the $1,000 to $2,500 band that holds about half of listings.