Most small ecommerce businesses treat product video like a social media project.

That made sense when the video lived on Instagram, TikTok, or YouTube and the product page did the selling later. But that line is getting thinner. Google added a new video_link attribute to Merchant Center in its 2026 product data specification update, and Google says those submitted videos become eligible to serve, with policy and quality validation checks, starting June 30, 2026.

That is a pretty clear signal. Product video is no longer just a brand asset. It is becoming product data.

If you sell online, this matters because AI-assisted shopping needs facts it can inspect. A shopper might ask Google, Gemini, or another buying assistant, “Which compact stroller folds with one hand?” or “Show me waterproof work boots that do not look bulky.” A plain product photo and a vague description may not carry enough proof. A short video can show the hinge, the texture, the fit, the scale, the packaging, the sound, or the way the product works in a real setting.

You do not need a Hollywood shoot. You need clear, useful product evidence.

Google’s new video link documentation says the video_link attribute lets merchants provide more visuals beyond image attributes, including videos that show a product from different angles or demonstrate the product in use. Google also says each product can include up to 10 video URLs, and supported formats include MP4, MOV, AVI, WMV, FLV, MPEG-1, MPG, and MPEGPS.

The technical requirements are not optional if you want Google to use the video. Google says video length must be between 6 seconds and 240 seconds, the file must be no larger than 500 MB, the aspect ratio should be 9:16, 16:9, or 1:1, and resolution should be at least 720p. Google also says videos must be publicly accessible, crawlable by Googlebot, stable, and not blocked by robots.txt.

That last part is where small stores often trip. A product video embedded in a fancy app, hidden behind a script, or stored at a temporary URL may look fine to a human but fail as product data.

Google’s broader product data specification says inaccurate, incomplete, or conflicting product information can cause disapprovals, limited eligibility, incorrect displays, or Merchant Center issues. Google lists common problems such as incorrect categories, missing GTINs, incorrect variant attributes, low-quality images, and conflicts between feed data and the website.

The same mindset applies to video. If the video says one thing, the product page says another, and the feed says a third, you have not built trust. You have built friction.

Video is becoming part of the buying decision

This is not only a Google feed issue. Buyers already use video to make decisions.

Wyzowl’s 2026 video marketing research says 85% of people have been convinced to buy a product or service by watching a video, and 84% of consumers want to see more videos from brands in 2026. That does not mean every product needs a glossy commercial. It means customers want to see the thing before they trust it.

Adobe also reported that traffic from AI sources to U.S. retail sites grew 393% year over year in the first three months of 2026, and that AI traffic converted 42% better than non-AI traffic in March 2026. Even if your store is small, the direction is hard to ignore. More shoppers are arriving from AI-assisted research, and those shoppers may be deeper in the decision process when they land.

That makes your product page’s job harder and simpler at the same time.

Harder because the buyer has already compared options. Simpler because they usually need fewer clever slogans and more proof.

Baymard’s product page research library repeatedly focuses on the details shoppers rely on, including product images, product variations, descriptions, specifications, shipping and returns, reviews, and the buy section. Product video sits right in the middle of that. It can answer questions that photos and copy often leave open.

What a useful product video actually shows

A useful product video is not a brand montage. It is not a slow pan over packaging with music. For ecommerce, the video should reduce buyer uncertainty.

Start with the questions customers already ask before they buy:

  • What size is it in a real hand, room, bag, shelf, truck bed, or workbench?
  • How does it move, fold, stretch, close, connect, pour, clip, charge, clean, or install?
  • What comes in the box, and what does not?
  • What does the material look like under normal light?
  • Which product variant is being shown?
  • What problem does this solve better than the cheaper option?

If the product is apparel, show fit, movement, fabric thickness, pockets, closures, and size context. If it is a tool, show the grip, bit changes, setup, noise, dust, cuts, measurements, and finished result. If it is skincare, show texture, amount used, packaging, scent notes if relevant, and who the product is not for. If it is food, show scale, packaging, prep, serving size, storage, and allergen clarity.

The goal is not to entertain everyone. The goal is to help the right buyer say, “Yes, this is the one,” and help the wrong buyer leave before they order and return it.

The small business product video stack

You can build this with basic equipment. A recent phone, a tripod, good lighting, and a clean surface are enough for many products.

The better investment is planning.

Create one repeatable template instead of reinventing every shoot. For example:

  1. Show the product name and exact variant for two seconds.
  2. Show scale next to a hand, common object, model, room, vehicle, or workspace.
  3. Demonstrate the main action or use case.
  4. Show one close-up detail customers care about.
  5. Show packaging, included parts, or fit notes.
  6. End with a clear still frame of the product.

That template can produce a 30 to 90 second video for most products. It also keeps you inside Google’s 6 to 240 second video length requirement.

Do not start by filming every SKU. Pick the products where video can change the outcome:

  • Best sellers with lots of traffic but average conversion
  • Products with frequent returns or pre-sale questions
  • Products where size, fit, assembly, texture, or operation is hard to explain
  • High-margin items where one extra sale pays for the shoot
  • Products already getting impressions in Google Merchant Center but weak clicks

This keeps the project tied to money instead of vibes.

Feed it, do not just post it

Publishing the video on a product page is useful. Submitting it correctly is the part many stores will miss.

Google’s video_link rules say the URL can be a YouTube URL or a URL from another video hosting site, but if it is not YouTube, the URL must point directly to a raw video file rather than a landing page with a video player. Google also says the video URL must be stable, static, publicly accessible, and crawlable by Googlebot.

That means your workflow should look like this:

  • Add the video to the product page where buyers can see it.
  • Add the right video_link value to the Merchant Center product feed.
  • Make sure the product shown in the video matches the variant in the feed.
  • Check Merchant Center diagnostics after submission.
  • Keep the video URL stable when you redesign the site or change apps.

If you use Shopify, WooCommerce, BigCommerce, or a feed management tool, check whether video_link is supported yet. If it is not, you may need a custom feed field, supplemental feed, app update, or Merchant API workflow.

This is boring work. It is also the kind of boring work that separates a clean ecommerce operation from a store that keeps guessing why products do not show correctly.

Do not let video slow the page down

A product video that makes the page drag is not a win.

Google’s page experience guidance says Core Web Vitals measure loading performance, interactivity, and visual stability, and Google recommends providing a good page experience for users. Product videos can hurt that if you autoplay heavy files, load several embeds at once, or push the buy button down the page on mobile.

Keep the page fast:

  • Use a poster image instead of loading the full player immediately.
  • Lazy-load the video below the main product image.
  • Avoid autoplay with sound.
  • Compress the file without making the product hard to see.
  • Test the product page on a real phone, not only your office desktop.

A small store does not need a perfect Lighthouse score to sell. It does need a product page that loads before the buyer gives up.

What to measure after launch

Video should earn its keep. Do not judge it only by views.

Track the product page before and after adding video. Watch conversion rate, add-to-cart rate, return rate, support questions, Merchant Center diagnostics, Shopping impressions, and organic product page traffic. If the product has enough volume, compare similar products with and without video for a few weeks.

Also track what customers ask after the video goes live. If people still ask, “How big is this?” the video failed that job. If they ask fewer sizing or setup questions, it is doing useful work even before the analytics look dramatic.

For stores with limited time, create a simple monthly routine. Pick five products, film five practical videos, submit them correctly, then review results 30 days later. Repeat with the next batch.

A practical 30-day plan

Week one: choose 10 candidate products. Pull customer questions, return reasons, support tickets, reviews, and search terms. Pick the five products where a visual answer would matter most.

Week two: write short shot lists. Do not write scripts unless you need narration. For most products, captions and clear visuals are enough. Confirm the exact product variant, color, size, model number, and included accessories before filming.

Week three: film and edit. Keep each video focused. Use steady shots, clean lighting, readable close-ups, and no distracting music. Export in a Google-supported format, then confirm resolution, file size, and length against Google’s video link requirements.

Week four: publish the videos on the product pages, add video_link values to the feed, check Merchant Center diagnostics, and record the baseline numbers you will compare next month.

This is not glamorous. It is shop-floor marketing. Show the product clearly, submit the data correctly, and measure whether buyers trust it more.

When product video is not worth it

Not every product deserves video right now. If a product has low margin, little traffic, no confusion, and weak inventory depth, skip it. If your product data is broken, fix the feed first. If your product photos are poor, improve photos before adding video.

Video is not a shortcut around bad merchandising. It is a multiplier for clear merchandising.

The businesses that benefit most will be the ones that already know their products, customers, and common objections. The camera just makes that knowledge easier for buyers and machines to understand.

If your ecommerce site needs a cleaner product page system, better Merchant Center data, or a video-ready product template, we can help. Start here: /get-started/.