Shopify Video Analytics: Track Views, Clicks and Attributed Revenue per Reel

Shopify video analytics come down to four numbers per reel: how many people saw it, how many clicked a product tag, how many added that product to cart, and how much revenue those carts turned into. Shopify's own reports give you none of the four at video level. They come from the widget that served the video, and they are only as trustworthy as the attribution rule behind them.

That last part is what most guides skip. Two tools can show different revenue for the same video on the same day because one counts a sale after a tag click and the other counts it after a view. Neither is lying. They are answering different questions.

This guide covers the four metrics, where each one comes from, how attribution rules differ, and how to read per-reel numbers without drawing conclusions from noise. If you are still building the case for video on your store, the shoppable video ROI case study covers what the results looked like on one store over a full quarter.

The four numbers that matter per video

Every video on your store should report these four, and the ratios between them are more useful than any of them alone.

Views. The number of times the video was seen. Definitions vary by tool, as the next section covers, so treat this as a count of impressions rather than of attention.

Product tag clicks. The number of times a shopper tapped a product tag on the video. This is the first signal that the video did more than fill space. Clicks divided by views is the video's click rate, and it is the number most sensitive to the video itself rather than to the placement.

Add-to-carts. The number of times a shopper added the tagged product to cart from the video. Add-to-carts divided by clicks tells you whether the product card the tag opens is doing its job: price, variants, availability and the add-to-cart button itself.

Attributed revenue. The order value the tool credits to the video under its attribution rule. This is the number that justifies the time spent making the video, and the one most easily misread.

SwipeReel reports views, add-to-carts and attributed revenue per video, which is the minimum set for the analysis in this guide. Whatever tool you use, confirm it reports at the individual video level and not only as a feed total. Feed totals hide the one reel that is doing all the work.

Why Shopify's own reports cannot do this

Shopify analytics attribute orders to sessions, and sessions to a traffic source, a landing page and a marketing channel. That model works for "did Instagram send buyers" and for "did the summer landing page convert". It has no field for "which on-site video was watched before this order".

There are three partial workarounds, and each has a catch.

• Landing page reports. If a video only exists on one page, orders from sessions that landed on that page are a rough proxy. It breaks the moment the same video appears elsewhere or the shopper enters via the home page.
• Order tags or cart attributes. A widget can write a cart attribute such as the video ID when the shopper adds from it, and that attribute lands on the order. This is the most accurate route inside Shopify, but you have to export orders and group by attribute yourself.
• Product sales reports. If a video promotes one product, that product's sales before and after the video went live is a crude before-and-after. It cannot separate the video from anything else that changed.

The practical answer is to let the video app own per-video attribution and use Shopify's reports for what they are good at: total revenue, channel mix and product performance. When the two disagree on a number, the app's figure is a subset of Shopify's, never the other way round. If the app ever reports more attributed revenue than Shopify reports in total for that period, its attribution rule is too generous.

How attribution works inside a video widget

Attribution is a chain of events tied together by a session or a visitor ID. Understanding the chain tells you what a revenue figure includes.

1. The widget records a view when the video meets its view definition.
2. The shopper taps a product tag. The widget records the click and remembers which video it came from.
3. The shopper adds the product to cart from the widget's product card. The widget records the add-to-cart and, in most implementations, writes the video ID to the cart as an attribute or to local storage.
4. The order completes. The widget matches the order to the earlier add-to-cart via the attribute, the session, or a webhook, and credits the order value to the video.

The differences between tools sit in two places. First, what counts as the trigger: some only credit an order if the add-to-cart came from the video, others credit any order in a session where a tag was clicked, and a few credit any order after a view. Second, the window: same session, same day, or a number of days.

Add-to-cart from the video is the strictest common rule, and it is the one worth using for decisions. It undercounts a little, because some shoppers watch, leave, and come back later through the product page. That undercount is a small price for a number you can trust.

Setting up GA4 alongside the app

The app's dashboard answers "which video sold". GA4 answers "how did video change behaviour across the site", which the app cannot see. Both are worth having.

GA4's Enhanced Measurement includes a video engagement toggle, but it only listens to embedded YouTube players. Self-hosted video in a theme section and app-hosted video in a widget fire nothing by default.

Two ways to fill the gap:

• Events pushed by the app. Some video apps push events to the data layer or directly to GA4 when a video plays, when a tag is clicked and when an add-to-cart happens. Check the app's documentation for the event names and enable it if offered.
• Google Tag Manager triggers. GTM has a built-in YouTube trigger and, for HTML5 video, a community template or a small custom listener that fires on play, pause and ended events from video elements. Pair it with a click trigger on the widget's product tag element.

Send three events at most: a play event with the video ID, a product click with the video and product IDs, and an add-to-cart with both. Then build one exploration in GA4 comparing sessions that fired a play event against sessions that did not, on conversion rate and average order value. That comparison is the closest a store gets to a holdout test without running one.

Do not add UTM parameters to on-site video links. UTMs on internal links start a new session in GA4 and overwrite the original traffic source, so you gain a video tag and lose the answer to where the shopper came from.

Reading per-reel numbers without fooling yourself

Per-video conversion rates on a Shopify store are small numbers derived from small numbers, and they move violently. A reel with 40 views and one order shows a 2.5 percent conversion rate. The next order takes it to 5 percent. Neither figure means anything.

Four habits keep the analysis honest.

Judge placements and feeds before videos. A home page carousel aggregates every clip in it. A collection page feed does the same. Those totals reach a usable sample size in days. Individual reels take weeks on most stores. Decide where video works before deciding which video works.

Set a minimum view count before you rank videos. A few hundred views is a reasonable floor. Below it, sort by views to see what is being seen, not by conversion rate.

Compare like with like. A reel on a product page is seen by people who already chose that product. The same reel on the home page is seen by everyone. The product page version will always convert better, and that says nothing about the reel.

Watch the ratios, not the totals. A video with a high click rate and a low add-to-cart rate has a product card problem, usually price, an out-of-stock variant or a confusing option selector. A video with a low click rate and a high add-to-cart rate is doing its job for the few people it convinces; the fix is more views, not a new video. The conversion rate guide for shoppable video works through several of these ratio patterns with examples.

A weekly review that takes fifteen minutes

Analytics that nobody reads are decoration. This routine is short enough to keep.

Do it on the same day each week so the comparison is week over week and not skewed by a weekend. Once a month, add a longer look: which products keep appearing in the top reels, and whether the reels that sell share a format, a length or a source. Content imported from Instagram Reels and TikTok often outperforms polished studio clips on the same product, and the monthly view is where that pattern shows up. The Reels to Shopify sales guide covers how to build a feed from that finding.

Attribution mistakes that inflate the numbers

A few errors show up on almost every store the first time it reviews video revenue.

Double counting across tools. The video app credits an order to a reel. Meanwhile the email tool credits the same order to a campaign and the ads platform credits it to a click. Each is correct under its own rule. Add them up and you have counted one order three times. Never sum attributed revenue across tools.

Counting views that were never seen. A video below the fold in a carousel that autoloads its poster can count as a view in tools with a loose definition. Compare the widget's view count against the page's session count. If views are higher than sessions on a page with one video slot, the definition is too loose.

Judging a video by revenue when the product is out of stock. Clicks without add-to-carts often mean the variant sold out. Check stock before you conclude the video failed.

Ignoring the control. If overall conversion rate went up the week you added video, and so did a discount, the video did not do it alone. The GA4 comparison of sessions with and without a play event is the closest thing to a control you have. Use it.

Turning the numbers into decisions

The point of tracking is to change what you do. Three decisions the four metrics support well:

Where to put video. Rank placements by attributed revenue per thousand views. The winner gets more feeds and more videos. The loser gets removed, which also gives back a little page speed.

Which videos to keep. Above the view floor, keep the top half by click rate and replace the bottom quarter. A steady rotation beats a one-time upload.

What to film next. Look at the products and formats behind the top reels and make more of those. If unboxing clips from customers beat studio product spins, that is a brief for the next month, and the video commerce guide covers how to plan a content calendar around what the data shows.

None of this requires a data team. It requires a tool that reports per video, a rule you understand, and a fifteen minute habit.