Where to Embed Product Videos on Your Site (and Where Not To)

Most product videos don't fail because they're bad. They fail because they sit where nobody presses play, or where the viewer needs a different video. The same 90-second demo can carry a homepage and get ignored in a footer.

The short answer: embed your main product video above the fold, just under the headline, with the play button visible without scrolling. Put short, muted clips next to individual feature claims. Use a trust-building video near the pricing table, not a second product tour. Put a short how-to video inside the product right after signup. In email, use a thumbnail that links to a hosted video. Skip footers, sidebars and popups that open on page load.

Two pieces of evidence shape that advice:

• Wistia studied 95,000 pages with embedded videos, split into 250-pixel zones from the top. Play rate dropped sharply between the third and fourth zones, about where the fold sits on most screens (Wistia).
• In email, only about 24% of clients support the HTML video element at all. Gmail doesn't play it on any platform (Can I email).

The rule behind every placement below: ask what the visitor is deciding at that point on the page, then give the video that one job. One video can't do every job, so plan a few cuts of the same story.

Product Video Placement at a Glance

The lengths come from our guide to how long a product demo video should be, which covers the word budget for each.

If you only have one video today, give it the hero spot and plan the shorter cuts next.

Hero Section: Keep the Play Button Above the Fold

The hero is where play rate is highest, according to Wistia's position study. The fold matters more than the exact pixel. In Wistia's data, zones 1 and 2 (the top 500 pixels) were above the fold on most screens, and zone 3 only on some. Play rate fell off sharply after that.

Two ways to use the spot:

• Video under the headline. The headline and subhead explain the product, and the video proves it. The page still works for visitors who never press play. This is the safer default for most SaaS pages.
• Video replacing the hero. Higher impact, higher risk. If it loads slowly or opens on a logo, visitors see nothing useful.

Size matters too. Wistia found the highest average play rates for embeds 401 to 600 pixels wide.

Whichever you pick, the poster image (the still shown before playback) is doing most of the work. Use a frame that shows the product's result, not a title card. Nielsen Norman Group's video usability guidelines suggest the thumbnail can come from the middle or end of the video for exactly this reason.

Takeaway: check your hero on a 13-inch laptop and a phone. If the play button is below the fold on either, move it up or shrink the space above it.

Autoplay: Muted Loops Only, With a Pause Button

Autoplay with sound mostly decides itself. Chrome's autoplay policy says muted autoplay "is always allowed," but sound needs an earlier user interaction or a high engagement score with your site. So a first-time visitor's sound-on autoplay will often just not start.

Muted autoplay is still a choice to make with care:

• Make it short and silent by design. A 10-20 second loop of the product working, with on-screen text, works without sound.
• Add a pause control. WCAG 2.1 success criterion 2.2.2 Pause, Stop, Hide is Level A. It requires a way to pause any moving content that starts automatically, lasts over five seconds and sits next to other content.
• Add playsinline for iPhones. Google's web.dev guide to lazy-loading video lists autoplay, muted, loop and playsinline for an autoplaying loop, and notes playsinline is needed for autoplay on iOS.
• Keep the real demo click-to-play. NN/g found users "don't appreciate being surprised by video or audio content that begins playing without their consent."

Takeaway: autoplay the loop, never the narrated demo.

Feature Sections: One Short Clip per Claim

A feature section makes one claim, like "turn a PDF into a video." A 10-30 second clip that shows exactly that is stronger proof than a link to the full tour.

• Match the clip to the claim. If the copy says "edit by typing," the clip shows someone typing a change and the video updating.
• Cut clips from your long demo. You don't need new recordings. Our guide to turning one demo into five clips shows how.
• Lazy-load anything below the fold. web.dev recommends preload="none" plus a poster image for videos that don't autoplay. The page then doesn't download clips nobody has scrolled to.

Takeaway: if a feature clip could sit next to any section on the page, it's too generic. Recut it until it only fits one.

Pricing Page: Answer Doubt, Not "What Is It?"

By the time someone reaches pricing, they've usually seen the product. Another feature tour answers a question they've stopped asking. What they want to know now is closer to "will this be worth it, and how hard is setup?"

Videos that fit that job:

• A setup walkthrough showing how long it takes to get the first result
• A customer story. Our guide on turning a case study PDF into a testimonial video covers this.
• A plan explainer if your tiers are hard to compare at a glance

Place it next to or just below the plan table, where the doubt happens. Don't put it above, where it pushes prices down the page.

Takeaway: write down the top question your sales team hears about pricing, and make that the video's script.

After Signup: Put the Video Inside the Product

A signup isn't the win. The first real use is. A short video placed in the product, at the moment someone lands on an empty dashboard, catches users who never open your welcome email.

• Keep it under 2 minutes and about one task: the first thing a new user needs to do.
• Put it where that task starts, like an empty state or a first-run checklist. Don't put it in a help center they have to go looking for.
• Let users close it and find it again later.

Our tutorial video guide covers how to script and record this kind of video.

Takeaway: if your activation step has a drop-off, that's where your onboarding video goes.

Email: Link a Thumbnail, Don't Embed

"Embedding" a video in email almost never means playback. Per Can I email, Gmail doesn't support the video element on desktop, iOS or Android. Apple Mail supports it only partly, and doesn't support autoplay. Outlook.com doesn't support it.

What works instead:

• A static thumbnail with a play button, linked to the video's page on your site
• An animated GIF of a few seconds if you want motion. It still links out.
• Text that stands on its own, so the email works if images are blocked

Link to a real page, not a bare video file. A page where the video is the main content is also the kind Google can index as a video. Our video schema and sitemaps guide explains the "watch page" rule.

Takeaway: the email's only job is the click. The viewing happens on your site.

Where Product Videos Underperform

Some spots look fine in a mockup and do little in practice:

• Footers. Wistia's data shows play rate falling the further down the page a video sits. By the footer, most visitors have scrolled past or left.
• Popups that open on page load. Google's guidance on intrusive interstitials warns that dialogs covering content on arrival can hurt how your page performs in search, as well as frustrate users.
• Sidebars. They're easy to read as ads and get ignored. A small embed also misses the 401-600 pixel width where Wistia saw play rates peak.

A video can still work in these spots for a specific reason, like a sidebar on a long tutorial. Just don't put your main demo there.

Takeaway: if the video isn't tied to a decision the visitor is making right there, move it or cut it.

How to Test Your Placement

You don't have to guess which placement works. Measure it:

1. Play rate first. Plays divided by page loads. A low play rate is a placement or poster problem, not a video problem. Our demo video metrics guide shows how to calculate it.
2. Then conversion. Compare sign-up rates for visitors who watched against those who didn't. It shows a link, not cause, so confirm with a split test.
3. Move one thing at a time. Test the video above vs. below the fold, or the poster frame, but not both at once.

Poko Motion makes the different cuts quicker to produce. It builds a roughly 60-second walkthrough for a homepage, a 40-45 second motion ad for social, and a longer voiceover explainer from a URL, repo, deck or PDF you already have. You refine each one by typing instead of editing a timeline. It can't tell you where on your page each one belongs. The table above is for that.

Your Next Step

Open your homepage on a laptop and a phone, and note where your product video's play button sits. If it's below the fold, move it up this week. Then check your video host's play rate before and after, and give the change two weeks of traffic before you judge it.