What each analytics metric means

Two definitions used throughout: a video counts as watched after 3 seconds of playback, and the attribution window (1, 7 or 30 days) is how long after seeing a video we keep counting a shopper's purchases.

The headline card. It splits everyone who visited your store into two groups — those who watched at least one video and those who didn't — and shows what each group was worth over the same attribution window.

Total revenue divided by the number of shoppers in the group. This is the number to watch. It combines how often people buy with how much they spend, so a single large order can't distort it. Across Storista stores, video watchers are typically worth 3–5× a non-watcher.

The share of shoppers in each group who placed an order. This is usually where the biggest difference between the two groups shows up — video mainly changes how often people buy.

What a typical order was worth in each group, shown side by side rather than as a ratio. For most stores the two are close, because video changes purchase frequency more than basket size. A small difference in either direction is normal and isn't something to fix.

How many shoppers watched at least one video, and what share of the people who saw your widget that represents. A low share means your videos are being shown but not played.

Total revenue from shoppers who watched, and how much more that is than the same number of shoppers would have produced at your store's normal rate.

What this card is, and isn't. Shoppers choose whether to watch, so the two groups aren't identical to begin with — someone who stops to watch a product video is often already more interested. This card compares two groups of shoppers; it doesn't isolate what the video alone caused. It's the right tool for "are the shoppers engaging with my videos more valuable?", which is the question that matters when deciding where to place videos and what content to make more of.

Each step shows how many shoppers got that far, and what share of the previous step that represents.

The two gaps that matter: Widget loaded → Saw video content is a placement problem, meaning the widget is on the page but too far down to be seen. Saw video content → Watched a video is a content or thumbnail problem, meaning shoppers see it and scroll past.

How many times your video content was displayed to a shopper. One shopper scrolling past three videos generates three impressions. Unique impressions counts distinct shoppers instead.

The share of shoppers who loaded a widget and then opened a video from it. It measures how compelling your thumbnails and placement are, rather than the quality of the videos themselves.

The share of shoppers who loaded the widget and actually scrolled it into view. Low visibility means the widget is placed too low on the page.

Mobile and desktop behave differently: mobile shoppers are the large majority for most stores and open the fullscreen player noticeably more often. Use the All devices / Mobile / Desktop control to switch. A small Unknown slice means the device couldn't be detected for those events.

Your videos play in two different situations, and they behave very differently.

The video plays inside the widget as the shopper scrolls — usually muted, often briefly. It reaches many more shoppers and produces a good conversion rate.

The shopper deliberately tapped to open the player. Fewer shoppers get here, they watch far longer, and their orders are typically substantially larger.

Share of shoppers who saw a video and opened the fullscreen player.

Don't optimise in-page watch time. Long in-page watch time usually means a carousel kept playing while the shopper was doing something else — it doesn't predict more orders. Fullscreen watch time does. If you want one engagement number to improve, use the fullscreen open rate.

Watch time is total seconds watched. Average watch time is per play. Completion rate is how far into the video a typical shopper got — the best guide to whether a video is too long or loses people early.

Fullscreen and in-page watch time are recorded separately from the date playback tracking was added, so for older date ranges the two can add up to less than total watch time.

Shoppers can only add to cart from a video that has a product tagged to it. Videos without a tagged product can't produce add-to-cart activity, so they're excluded from the add-to-cart rate rather than counted as failures. The card shows how many of your videos are tagged.

SCREENSHOT PLACEHOLDERThe Add to cart card, including the "X of Y videos have products tagged" prompt

Add-to-cart clicks counts every tap. Unique shoppers counts distinct people. Add-to-cart rateis clicks per shopper who saw a tagged video.

Buy-now taps skip the cart and go straight to checkout, so they're counted separately rather than mixed into add-to-cart.

This card counts only orders we can credit to a specific widget or video. It answers a narrower question than the watcher card at the top of the page.

Most video-driven revenue is influenced rather than direct, and that's expected. Direct only happens when a shopper uses the in-video cart button, which is a small share of how people shop.

Why this doesn't match the card at the top of the page. The watcher card counts everything shoppers who watched went on to spend, wherever they bought it. This card counts only orders traceable to a particular widget or video. Both are correct — they answer different questions. Use the watcher card to judge whether video is working, and this card to judge which widget or video deserves credit.

Top videos ranks your content across every widget it appears in, so a video used in three places is listed once with its totals combined. Sort by watch time to find what holds attention, by CTR to find what earns the click, or by revenue to find what sells.

Top widgets compares placements rather than content — the same video can perform very differently in a homepage carousel and on a product page.

SCREENSHOT PLACEHOLDERTop videos table and Top widgets table

A dash (–) in a table means there's nothing to report for that row yet. It is not the same as zero.

Opening a widget shows the same metrics scoped to that placement, plus a per-video Posts breakdown with impressions, CTR, fullscreen and in-page watch time, completion rate, add-to-carts and revenue for each video in that widget.

SCREENSHOT PLACEHOLDERWidget detail page with the Posts breakdown table