Shoppable Video in Email and SMS: What Actually Works
Almost no inbox will play an embedded video, so the thing that works in email is an animated poster frame linking to a feed. This is the pattern, the client-by-client reality, the MMS size ceilings, and how to attribute the sale to the clip rather than the campaign.
On this page
Key takeaways
- Do not embed video in email. Apple Mail is the only mainstream client that plays HTML5 video reliably; everywhere else your subscriber sees the fallback, so design the fallback as the real thing.
- The pattern that works is an animated poster plus a deep link: a short GIF or still with a play affordance, linking to a video feed rather than a static product page.
- Keep the GIF under about 500KB and put the message in the first frame, because Outlook on Windows shows only the first frame of an animated GIF.
- For SMS, send a link, not a file. Twilio documents a 5MB ceiling for MMS and 16MB for RCS, and carrier handling below those numbers is inconsistent.
- Attribute to the clip, not the campaign. Put the clip identifier in the landing URL so your video analytics and your email platform agree on which creative earned the order.
Shoppable video does not really work inside an email. What works is an animated poster frame that looks like a video, linking to a feed where the video actually plays and the products are actually tagged. Nearly every inbox strips or ignores an embedded player, so the thing your subscriber sees is the fallback, which means the fallback is the design.
SMS is simpler and stricter: send a link. The file-size ceilings on MMS are low enough, and carrier behaviour inconsistent enough, that pushing video down the wire is a worse experience than a short message pointing at a feed that loads in a second.
This post covers what each channel can actually render, the pattern to build instead, and the piece most teams skip, which is attributing the order to the specific clip rather than to the campaign it was in.
Why video in email is almost always a GIF or a poster frame
HTML5 video in email has been technically possible for over a decade and practically unusable for just as long. The <video> element is supported by a small minority of clients, and the ones that matter commercially mostly are not in it. Gmail strips the tags. Most Outlook builds block playback through content security policy. iframes and script embeds, which is how you would normally put a YouTube player anywhere, are stripped in essentially every client.
What this means in practice is that the poster attribute, not the video, is what your audience sees. So the sensible design move is to stop treating the poster as a fallback and start treating it as the creative. Design the still frame to carry the message, add a play affordance so the reader understands there is video on the other side of the click, and spend your effort on where that click lands.
There is one real advantage to this that gets overlooked: a poster plus a link is identical in every client. You test it once. An embedded video is four different experiences across four clients, and you will spend the whole QA cycle on the 5% of your list that might see the good one.
Client support changes and vendor charts go stale
Client-support tables for email video are published mostly by email tool vendors, they disagree with each other, and many are years old without being dated. If your campaign depends on a specific client rendering something, test it yourself in that client this week rather than trusting a chart. The one claim that has held consistently across sources and years is the broad one: assume the embedded video will not play, and build for that.
The clients that will and will not play an embedded video
Treat the table below as a planning assumption rather than a test result, and treat anything in the middle column as something you should verify against your own list's client breakdown before you build around it.
Embedded HTML5 video in email, by client
| Client | Plays embedded video | What the subscriber sees instead | Plan for |
|---|---|---|---|
| Apple Mail (macOS, iOS) | Generally yes | Plays, though iOS tends to ignore autoplay and needs controls | The only place a real embed pays off |
| Gmail (web and mobile) | No | The poster image, if you set one | Poster and link |
| Outlook on Windows | No | Poster image; animated GIFs show only the first frame | Static first frame carrying the message |
| Outlook.com and Outlook mobile | Reports conflict | Usually the poster image | Poster and link |
| Yahoo, AOL, Samsung Mail | Reports conflict | Usually the poster image | Poster and link |
The Outlook-on-Windows row is the one that quietly ruins campaigns. It renders only the first frame of an animated GIF. If your GIF's whole idea is a reveal at second three, a large share of your B2B list sees a shot of nothing. Put the message in frame one and let the animation be a reward rather than the payload.
Animated poster plus deep link: the pattern that actually converts
Here is the build, in the order you should do it.
Cut a three to five second loop from the clip itself
Not a bespoke animation. The point is continuity: the subscriber should recognise the moment when they land on the feed. Export it as a GIF at the width you are actually displaying it, and keep it under about 500KB. Over that and you are adding load time to an email, which is a strange place to be slow.
Compose frame one as a complete still
It has to carry the offer and the product on its own, because for Outlook on Windows that frame is the entire creative. Add a play triangle. It is a visual convention strong enough that readers understand it means "there is video here" even in a still image, and it measurably raises click intent compared to a bare photo.
Link it to a feed, not a product page
This is the step that makes it shoppable rather than decorative. A subscriber who clicked a video thumbnail has told you they want to watch something. Landing them on a static product page breaks that promise and wastes the intent. Land them on a swipeable feed or a product page with the video feed in view and autoplaying, so the thing they clicked for is the thing they get. The phone-shot product video workflow gives you clips that are already the right shape for this; vertical video makes a better email thumbnail than a wide crop does.
Keep the whole email to one clip
Three video thumbnails in one email is three times the weight for a reader who will click at most one. Pick the clip, build the email around it.
The email video build, start to finish
- Export a 3-5 second loop from the actual clip, under 500KB, at display width
- Compose frame one to work alone, with the offer and product visible
- Overlay a play triangle so the still reads as video
- Add alt text that describes the product, not the format, for images-off readers
- Link the whole image block, not just a button underneath it
- Point the link at a video feed, with the clip identifier in the URL
- Send a seed test to one Outlook on Windows address and look only at frame one
- Check total email weight; under 100KB of HTML plus the GIF is a sensible ceiling
SMS and MMS: file size ceilings and carrier behaviour
The temptation with MMS is to send the video itself. Resist it, for three reasons.
First, the ceilings are real. Twilio, which is the plumbing under a large share of ecommerce SMS, documents a maximum of 5MB for MMS and 16MB for RCS, and notes that if an RCS message falls back to MMS it rejects anything over the MMS limit. Five megabytes sounds like plenty until you remember that is roughly four seconds of the 1080p export your editor just handed you.
Second, what actually arrives varies. Carrier-specific limits for non-image files sit below the platform ceiling and differ by carrier and by route, and video is routinely recompressed somewhere in the chain. You do not control the quality of the thing your customer sees, which is an odd position to be in after spending a day on colour grading.
Third, and most practically, MMS costs more per message than SMS. At list scale that difference is not trivial, and you are paying it to deliver a lower-quality version of a video you are also hosting properly on your own site.
The alternative is boring and works. Send an SMS with one line of copy and a short link to the feed. The link preview on most modern handsets will pull your Open Graph image, which gives you the thumbnail anyway, for free, at full quality, with no size limit and no carrier compression. Make sure your feed page has a proper og:image and og:video set, because that preview is doing the work the MMS attachment was supposed to do.
If you do send MMS, send a short GIF or a still rather than a video file. It degrades more gracefully and it sidesteps most of the codec variability.
Landing the click on a video feed instead of a static PDP
The landing experience is where most of the value in this channel is won or lost, and it is the bit email teams tend to hand off and stop thinking about.
A shopper arriving from an email thumbnail has a specific expectation: the clip they saw, playing, now. The best version of that is a feed that opens on that exact clip, autoplays muted, and has the tagged product card visible without a scroll. The worst is a collection page where they have to go hunting for the thing they clicked.
Deep-linking to a specific clip is a feature, not a given, so check your app supports it before you design a campaign around it. If it does not, the workable fallback is a dedicated landing page per campaign with the feed ordered so the hero clip is first.
A few details that matter more than they sound. Make sure the feed is above the fold on a phone, because essentially all of this traffic is mobile. Make sure the video does not require a tap to start, because a shopper who clicked through from an email has already made one tap too many. And make sure the product card is the thing immediately under the video rather than a row of related products, since the entire point of the journey is the item in the clip. Our guide to Shopify marketing tactics covers how this slots into the wider campaign calendar.
Attributing email and SMS revenue to the clip, not just the campaign
Your email platform will happily tell you that the campaign drove revenue. That is a useful number and it is not the one that improves your video.
What you want to know is which clip earned the click and which clip earned the order, because that is the feedback loop that makes the next shoot better. Getting it requires one deliberate step at build time: put a clip identifier in the landing URL, as a query parameter, and make sure it survives into both your analytics and your video app's own reporting.
A URL pattern that works looks like your feed page plus the campaign UTMs plus a parameter naming the clip. Keep the naming scheme boring and consistent: the shoot date and a two-word description beats a hash nobody can read in a report six weeks later.
Then reconcile two numbers monthly. Your email platform's attributed revenue for the campaign, and your video app's attributed revenue for that clip across all sources. They will not match, and the gap is informative: a clip earning far more in the video tool than the campaign suggests is a clip that is also working on-site and deserves a better placement. A clip that only ever converts from email is usually one carrying an offer rather than a product story.
The mechanics of wiring this up properly, including what to do when two tools both claim the same order, are in our guide to Shopify video analytics and revenue attribution. The short version: pick one system as the source of truth for the headline number, and use the other for ranking creative.
One honest caveat on all of it. Email and SMS traffic is the most pre-qualified traffic you have. A clip that converts well from a campaign is partly just benefiting from an audience that already likes you. Compare clips against each other within the channel, not against your sitewide numbers, or you will conclude that every video is a triumph.
Frequently asked questions
- Can you embed a playable video in a marketing email?
- Technically yes, practically no. The HTML5 video element is supported by a small minority of email clients, with Apple Mail on macOS and iOS the main one that plays it reliably. Gmail strips the tags, and most Outlook builds block playback. Because iframes and scripts are stripped nearly everywhere, you also cannot embed a YouTube or Vimeo player. The working approach is to treat the poster frame as the creative and link it to a page where the video actually plays.
- Does Gmail support HTML5 video in email?
- No. Gmail does not play embedded HTML5 video in either its web or mobile clients; it strips the markup and your subscriber sees the poster image if you set one, or nothing if you did not. Gmail will generate a preview card for a YouTube link in some contexts, but an embedded player will not work because embedding requires JavaScript, which no mainstream email client executes. Given Gmail's share of most consumer lists, this alone is reason enough to design for the fallback.
- What size should an email video GIF be?
- Under about 500KB, three to five seconds, exported at the width you are actually displaying it rather than scaled down in the HTML. The bigger constraint is not size but the first frame: Outlook on Windows displays only frame one of an animated GIF, so that frame has to carry the offer and the product on its own. Build the still first, then let the animation be an enhancement for the clients that show it.
- Is MMS video worth the extra cost per message?
- Usually not. Twilio documents a 5MB ceiling for MMS and 16MB for RCS, carrier limits for non-image files sit below that and vary by carrier and route, and video gets recompressed in transit so you do not control what arrives. You also pay more per message than SMS. A plain SMS with a short link to your video feed costs less, delivers full-quality video, and on most modern handsets generates a thumbnail preview from your page's Open Graph image anyway. If you do use MMS, send a GIF or a still rather than a video file.
- How do I track which email clip drove the purchase?
- Put a clip identifier in the landing URL alongside your normal campaign UTMs, and make sure it reaches both your analytics and your video app's reporting. Your email platform attributes revenue to the campaign, which tells you the send worked but not which creative did. Reconcile the two monthly: a clip that earns far more in your video tool than the campaign number implies is also working on-site and deserves a better placement, while a clip that only converts from email is usually carrying an offer rather than a product story.
- Should the email link go to the product page or the video feed?
- The feed, or a product page with the feed in view and already playing. A subscriber who clicked a video thumbnail has signalled that they want to watch something, and landing them on a static product page breaks that promise at the exact moment their intent is highest. Ideally the feed opens on the specific clip they clicked, which requires deep-linking support, so check your app offers it before you design a campaign around it. If it does not, build a campaign landing page with the hero clip ordered first.


