Do product videos actually sell?

Most video statistics come from surveys of marketers who already use video, which makes them hard to separate from enthusiasm. The cleanest test we found is a randomized field experiment published in Management Science. Researchers added videos showing apparel styled with matching accessories to a few randomly selected items on a fashion retailer's website, and introducing a video raised sales of the apparel by 14.5% and of the accessories shown alongside it by 28.3% ¹. The authors attribute most of the accessory lift to spillover, meaning the video taught shoppers which products belong together, which is a useful hint about what a product video should actually show.

Crowdfunding offers a second, much larger dataset. Ethan Mollick's study of more than 48,500 Kickstarter projects found that, for the average project with everything else held constant, not including a video decreased the chance of success by 26% ². This is a correlation rather than an experiment, and Mollick treats the video as a signal of how prepared a founder is, so the lesson is less that video is magic and more that buyers read its absence as a lack of effort.

Survey data points the same way but deserves a discount. In Wyzowl's 2026 survey of 266 respondents, 83% of video marketers said video has directly increased sales and 82% said it gave them a good return, down from 93% the year before ³. Yet only 32% said they measure that return through bottom line sales, while 67% measure it through views ³, and Wyzowl sells video production. On the buyer side of the same survey, 85% of people said a video had convinced them to buy a product or service at some point ³, which tells you video is part of how people shop rather than how much it moves your particular numbers.

How the video is made seems to matter as much as whether it exists. A study in the Journal of the Academy of Marketing Science compared demonstrations that walk through the steps of using a product with ones that only show the finished result, and across five studies the step-by-step versions consistently produced higher purchase intentions, particularly when watched alone, the way most on-demand online videos are ⁴. Put together, the evidence supports a product video that clearly shows the thing working, measured against your own sales rather than against someone else's survey.