Video Retention Statistics: Retire the 95% Figure Before It Sinks Your Business Case
Quick answer:Video retention statistics such as 95% retained from video versus 10% from text trace to a late 2000s vendor survey of around 200 B2B buyers, and its publisher calls the result directional. Cite your own before and after baseline or a clearly attributed study, and drop any figure you cannot trace to a source page.
Where does the 95% video retention statistic come from?
Somebody on your finance team will click the footnote. That is the moment a video business case either holds or collapses, and the 95% figure is the one most likely to collapse it.
Insivia, the agency behind the number, says it came from a small survey of around 200 B2B buyers in the late 2000s that compared how well people recalled information from short video against written text. The agency's own write up says the percentages showed a pattern in its survey and were not universal laws of cognition, and it calls the result directional rather than a peer reviewed study. The figure then spread through slide decks with those caveats stripped off.
Which other video statistics fail a source check?
Three more figures show up in the same decks: that 73% of employees prefer leadership video, that video drives 75% higher engagement, and a claimed Forrester finding that people are 75% more likely to watch. We searched for a named primary source for each and found only aggregator blogs. We could not trace them, so we do not use them.
| Claim | What we found | Safe to cite? |
|---|---|---|
| 95% retained from video, 10% from text | Vendor survey of around 200 B2B buyers, late 2000s; publisher calls it directional | Only with that framing |
| 73% prefer leadership video | No named primary source found | No |
| 75% higher engagement | No named primary source found | No |
| Forrester: 75% more likely to watch | No Forrester source found | No |
| 38.3% average adoption increase | Forrester Consulting for Intellum, 2024; 300 customer education decision makers; commissioned and self reported | Yes, with attribution and the caveat |
What can you cite in a video business case instead?
Cite the studies you can open. The Forrester Consulting survey for Intellum is self reported and commissioned, and it is still a named study with a sample size you can state.
Capsule's published figures are 93% lower cost per video, 10x more videos, and 8x faster production. Those describe production, meaning what it costs and how fast teams make video. They say nothing about whether viewers remember more, so keep them in the production section of the case.
How do you build a baseline a finance partner will accept?
Pick one action a viewer can take: an enrollment, a form, a rollout step. Measure how many people take it after the text version. Then run the video against the same audience and compare.
Say an internal comms lead sends a benefits reminder as text to one region and as a short video to a comparable one. Compare the action rates. For more on why raw engagement misleads, seevideo volume and engagement, andperformance benchmarks that aim at the wrong target. Comms teams building the case will find more in ourinternal comms use case.
How should a pitch handle statistics it cannot verify?
Put the source and its caveat in the footnote, and put the claim in the body only if it survives the caveat. A finance partner who reads a footnote saying survey of around 200 buyers, publisher calls it directional will forgive the number. The same person who finds no footnote at all will start doubting every other line.
Keep an evidence log for the deck: the claim, the publisher, the year, the sample, and the link. It takes ten minutes to build and it answers the first question anyone asks in the review. It also keeps the deck honest when someone new edits it and pastes in a figure from a blog post.
Finally, ask the skeptic what number would convince them. Finance partners often have one in mind, such as a completion rate, a response rate, or hours saved. Measuring that number, even on a small pilot, beats quoting a famous statistic that nobody in the room believes.
FAQ
Is the 95% video retention statistic true?
It comes from a small vendor survey of around 200 B2B buyers, and its publisher describes it as directional. It is not a reliable finding about how much people retain.
What video statistics are safe to use in a business case?
Use figures with a named publisher, a sample size, and a page you can open, and state the caveats. Your own before and after baseline is safer than any borrowed number.
Why do so many video statistics have no source?
Figures get copied between blog posts, and each copy drops the original citation. After a few rounds, nobody can say who ran the survey.
How do you prove video works better than text internally?
Choose one measurable action, test the text version first, and run the video against the same kind of audience. The comparison is small, but it is yours.