What Is a Video Edition Test?
A video edition test publishes a video version of an existing audio podcast for a set period, so that its effect can be measured against the period without it. The edition can be filmed, or built from the audio, as Podnews’s was: Podnews asked for no generative AI, so Liquid Studios filled gaps from an image library, working from the script, audio, captions and images Podnews shared in a folder (Podnews). Our October explainer on AI-made podcast video editions covers the vendor’s pricing and claims when the pilot began; these are the first results.
How to Run a Video Test at a Glance
| Step | Action | Expected result |
|---|---|---|
| 1 | Record a baseline for the five weekdays before | Normal views, plays and downloads |
| 2 | Decide the formats and the deadline each day | Video live at the same time as audio |
| 3 | Publish every episode for the test week | A consistent run, not one-off uploads |
| 4 | Track each platform separately | YouTube, Spotify and downloads in their own columns |
| 5 | Keep measuring for a week after | Whether the lift lasts |
| 6 | Compare the gain with the cost | A keep-or-stop decision |
Time: two weeks of tracking. Tools: YouTube Studio, Spotify for Creators, your host’s statistics. Cost: production time or a vendor fee. Skill level: intermediate.
What You Need Before Starting
You need a YouTube channel already set up for the show, as described in our guide to creating a YouTube channel for a podcast, plus access to Spotify’s and your host’s numbers. Write down a stop rule before day one, such as “continue only if YouTube views stay at least double the baseline in the week after.”
Step 1: Record a Baseline
Copy daily YouTube views, impressions, engaged views and average view duration, plus Spotify plays and host downloads, for the five weekdays before the test. Podnews used that window. Note that YouTube changed how it counts on August 24, 2026: a view now counts “from the very first frame,” and the older definition, “clicking to watch or watching past the initial seconds,” is now called an engaged view (YouTube Official Blog). Compare like with like on either side of that date.
Step 2: Pick Formats and a Daily Deadline
Choose what you will publish, and make the video go live with the audio. Podnews ran a horizontal version, a vertical cut, clips and, from day two, chapter markers. It publishes at 4 a.m. Pacific, and the horizontal video arrived by 5 a.m. each day (Podnews).
Step 3: Publish Every Episode for the Whole Week
Do not skip days. Podnews treated launch day separately because it had a press release behind it, then averaged the other four. Do the same with any day you promote, or the average will flatter the format.
Step 4: Track Each Platform in Its Own Column
Keep YouTube, Spotify and downloads apart, because they answer different questions. Video was carried by the platforms, not the feed. When we checked the 99 public feeds behind Apple’s US top 100 on November 10, none had a video enclosure or video alternate enclosure in its latest 10 episodes. So a feed’s download count will not show video growth, and Podnews warns some “audio downloads” may be video plays depending on the app. Our explainer on whether Podtrac counts YouTube views like YouTube explains why the two numbers never merge cleanly.
Multiple of the comparison days (1 = no change)
Source: Podnews, “What happened when the daily Podnews podcast became a video”, November 10, 2026. Read November 10, 2026.
Podnews compared the pilot with the five weekdays before it and a few days after. It gave impressions as a median increase of 121%, shown here as 2.21x, and described views per edition as “almost” 4x and Spotify plays as flat. One show, one week.
The lift was confined to the platform that got the new format: YouTube multiplied, while Spotify, the other place listeners could find the show, did not move.
Step 5: Keep Measuring After the Test
Track the same metrics for at least a week after the last video. At Podnews, daily YouTube views went back to their previous level (Podnews). Editor James Cridland wrote: “Had we kept at it for longer, we might have achieved more of an algorithmic effect.” That is the open question a one-week test cannot answer.
Step 6: Compare the Gain With the Cost
Put the extra views next to what they cost and what they earn. Podnews paid nothing for the pilot and still stopped, because its podcast is “something that doesn’t specifically pay its way”: the show had 43,622 downloads in October against more than 675,000 newsletter sends (Podnews). By our arithmetic, that is about 1,400 downloads a day. A show that sells video sponsorships would weigh the same lift very differently.
How Do You Verify It Worked?
The test worked if you can state each metric’s change against the baseline and the after-period, platform by platform. Podnews’s summary is a good model: one line for YouTube, one for Spotify, one for downloads, one for what happened afterward.
Common Problems and Fixes
| Problem | Likely cause | Fix |
|---|---|---|
| Big jump on day one only | Launch promotion | Report day one separately |
| Views up, watch share down | Longer videos | Compare average view duration in seconds |
| Downloads unchanged | Video lives on platforms, not the feed | Judge video on platform metrics |
| Baseline not comparable | YouTube’s August 2026 view change | Use engaged views for older periods |
What Not to Do
Do not add YouTube views to audio downloads in a media kit. Consequence: a sponsor double-counts listeners. Correction: list each platform on its own line.
Do not judge a format on one week. Consequence: you mistake novelty for growth. Correction: decide the test length and stop rule in advance, and say what the test could not show.
Frequently Asked Questions
Does adding video grow a podcast?
In Podnews’s one-week test in October 2026, YouTube views rose sevenfold but Spotify plays and audio downloads did not change, and YouTube views returned to normal after the test. One show’s week is not a general rule.
How long should a podcast video test run?
Podnews ran one week and compared it with the five weekdays before. We suggest at least that, plus a week of tracking afterward, and a stop rule written down before you start.
What metrics should I track in a podcast video test?
YouTube views, impressions, engaged views and average view duration, Spotify plays and impressions, and host downloads, each in its own column for the baseline, test and after periods.
What is an engaged view on YouTube?
Since August 24, 2026, YouTube counts a view from the first frame and calls its older definition, clicking to watch or watching past the initial seconds, an engaged view, according to the YouTube Official Blog.
Do video plays count as podcast downloads?
Usually not. In our November 10, 2026 check, none of the 99 feeds behind Apple’s US top 100 carried a video file in RSS, so most video plays happen on platforms and appear in their own analytics.
Did the Podnews video pilot use AI?
Podnews says Liquid Studios assembled a first cut automatically and a human editor checked it. Podnews asked for no generative AI, so library images filled gaps.
Sources and Methodology
This guide is built from Podnews’s November 10, 2026 report on its video pilot, YouTube’s August 19, 2026 post on view counting, and our own check of the RSS feeds behind Apple’s US top-100 chart on November 10, 2026. The baseline method follows Podnews’s; the stop rule and tracking layout are our recommendations. We did not see Podnews’s raw analytics.
- trade press Podnews, What happened when the daily Podnews podcast became a video. All pilot figures, formats, timings and conclusions. Read November 10, 2026.
- platform documentation YouTube Official Blog, Engaged views explained. View and engaged view definitions, August 24, 2026 change. Read November 10, 2026.
- official Apple, Apple Marketing Tools RSS feed generator and Search API. US top-100 chart and feed addresses for our video check. Read November 10, 2026.
- live read Media Morpheus Research, scan of the latest 10 items in each of 99 public feeds for video enclosures and podcast:alternateEnclosure video types, November 10, 2026.