Skip to main content
Back to Blog
Analytics

Video Engagement Analytics and What to Measure

November 10, 202611 min read
Video Engagement Analytics and What to Measure

The five video engagement metrics to start with are views, unique viewers, average watch time, average watch percentage, and call-to-action clicks when your video has a button. They help you see how many people you reached, how much they watched, and whether they took the next step you offered.

Suppose you share a demo that shows how to export an expense report. You want people to reach the moment when the finished file appears. How you read the numbers depends on whether viewers are getting to that moment and what you want them to do afterward.

Open the analytics for one video

In Zidi, open the video and select View Analytics below the player. Video analytics are included on Free, Starter, and Pro. The report opens for the video you selected.

Choose the last 24 hours, last 7 days, last 30 days, or all time. For a window with a set length, the summary cards compare it with the previous period of the same length. The report uses UTC, so use that time zone when matching a viewing spike to the hour you sent a link.

The report filters out your own signed-in views and known bots. Published video webpages sit outside video-view tracking. For viewing activity from a campaign’s personal links, open that campaign’s analytics.

If you want to review several videos side by side, export the CSV for the period you chose on each video. Keep a note of when and where you shared each link. An expense report demo sent to a sales list on Monday has a different audience from a training recording people open as needed.

Understand views and unique viewers

Total Views shows how many views Zidi counted. Each browser contributes at most one view for that video in a 24-hour window, even if someone plays it several times.

Unique Viewers counts distinct browsers for anonymous viewers and distinct accounts for signed-in viewers. An anonymous viewer who watches on a phone and a laptop can count twice. Read the figure as a measure of reach using those counting rules.

Use the two numbers together. If Total Views grows while Unique Viewers barely moves, look at how the link was shared over the period. A team may be using a training video as a reference and coming back to it on later days. A prospect who returns to a demo later can also add a counted view.

When both numbers are low, check who received the link and where it appears. For the expense report demo, check that the message reached people who actually need to export reports. A link buried below several unrelated points in an email gives even the right person a reason to miss it.

The counting rules also matter when you compare reports from different tools. Check how each tool defines a view and a viewer before you put their numbers in the same summary. See the definitions of views vs unique viewers.

Read watch time and watch percentage together

Avg Watch Time is total play time divided by unique viewers. It tells you how much time the video drew per viewer. Read it alongside the length of the recording and the timestamp of the step you want people to see.

Avg Watch Percentage is the average share of the video actually played per session. Each portion counts once toward coverage in that session. If a viewer plays the opening and jumps to the export step, the skipped middle stays outside that session’s coverage.

A short walkthrough can have a high watch percentage even when it takes little time to watch. For a longer recording, check the time viewers spent against where the main task appears. If people spend a while with the video and the played share is low, use retention to see how far they got.

Consider a demo where the export happens late. People might spend time listening to the setup and leave before the file appears. Look at the point where viewers stop and decide whether the setup earns its place before the demo.

You can also use both numbers to evaluate a revision. If you move the export earlier, compare similar sharing windows and look at how many viewers reach the new timestamp. A shorter average watch time can make sense if viewers get the answer sooner. For the metric definition, see watch time.

Find where viewers stop watching

In the Retention tab, use Furthest Point Reached to see the share of viewers who got to each point. Then look at Key Drop-off Points, which shows up to four places where more than 10% of viewers stopped. Play the recording from a few seconds before each flagged point.

Check what the viewer is seeing and hearing at that moment. A drop just as you switch from the report screen to a long explanation might tell you where the walkthrough loses momentum. A drop while a menu is open might mean the next click or the label is hard to follow.

For the export demo, start with the first dip before the finished file appears. If the opening spends time on why the report exists, try showing the file first. Then work through the clicks that produce it.

When you revise a video, choose one moment to improve and compare that moment with the next batch of viewing activity. The timestamps may move if you cut the opening or move a step. Compare the same action in the demo, even if it now appears earlier on the timeline.

A viewer can reach a point by seeking ahead. On Pro, use the heatmap to see which parts people played around that point. See how to read a retention curve.

Use the engagement heatmap on Pro

The Engagement Heatmap shows which parts of a video more viewers played. Brighter sections were played by more viewers, and skipped sections stay cool. On Pro, use it with the retention view to see where people spent their attention.

Suppose the section where you click Export is bright, and the explanation before it is cooler. Play those two sections yourself. Ask whether the explanation tells viewers something they need before the click or whether it fits better after they see the result.

Use the heatmap when two sections seem to tell different stories. A point can be far along the timeline and still get played by viewers who jump to it. That can be useful for a reference video, where someone may already know the early steps and want help with one menu.

Write down the part you want viewers to play before you share the video. For the expense report demo, that could be the few seconds from opening the export menu to seeing the file. After sharing, check that span in the engagement heatmap. For plan details, see pricing.

Check traffic sources and audience

The Traffic Sources tab shows where Zidi detected the view coming from. You may see labels such as Direct, Google, YouTube, LinkedIn, Slack, or Email Client. Other websites can appear by domain name.

Some apps open a link without passing along where it came from. In that case, Zidi shows the visit as Direct. If you shared a video in a channel that often opens links this way, keep a note of when you posted it so you can interpret a rise in Direct views.

A UTM tag is a label you add to a share link to describe where you distributed it. Zidi uses the tag in place of the detected source. A link tagged with utm_medium=email appears as Email, utm_medium=cpc appears as Paid, and a social source appears as Social.

Set the tags when you prepare the link for each channel. If the expense report demo goes into an email and a social post, label the links for those uses. When you review the report later, the source names give you a way to connect the video’s views with where you sent people.

Check the Audience breakdown for desktop, mobile, and tablet viewing, plus the top countries. If many viewers are on mobile, watch the export step on a phone and check that the file name and button label are readable. The geographic breakdown is at country level.

See which prospects watched

The per-video report gives you a view of the audience as a whole. When your next step depends on a specific person, give them a tracked link or send a campaign with a personal link for each recipient.

A tracked link is a personal link tied to one email address. It shows whether the link recorded a view and how many times. That can be enough for a sales rep who wants to see whether a prospect opened the demo they asked for.

Use a campaign when you need more detail for each person. Campaign results show views, watch percentage, and whether the recipient clicked the call-to-action button when one is present. The campaign’s per-recipient watch percentage is the share of the video that person actually played. The video-wide average is the same kind of measure, averaged across viewing sessions.

For the expense report demo, one buyer might want to know if the export fits a monthly reporting process. If their watch percentage shows they played most of the demo, follow up with a question about who uses the file or where it goes next. Keep the conversation focused on the buyer’s workflow.

On Pro, Slack can alert you about a view or a call-to-action click. It also posts watch milestones for tracked-link recipients, but not for campaign recipients. Also on Pro, a connected CRM, the system your team uses to keep prospect and customer records, receives a watch note when a tracked-link recipient passes 50%. Zidi updates that note at 80% and 100%. For a campaign recipient, the CRM gets an opened note only if they open the page without pressing play, and a completed note when they reach 95% or click the call-to-action.

See the available integrations.

Decide what a call-to-action click means

A call to action, or CTA, is the next step you ask the viewer to take. On Pro, you can use an interactive CTA button. A sales demo can send someone to a calendar to book a walkthrough, and a training video can send them to the next lesson.

Read the click in the context of the ask. If the button opens your calendar, look there for a booking before you follow up. A click gives you a reason to check whether the viewer completed the next step and what they might need from you.

Use one clear action that fits the video. If the demo shows a buyer how to export a report, a button that offers a walkthrough of their own reporting process continues the conversation. The label should tell them what will happen when they click.

If few people reach the moment where you explain the next step, fix that part of the recording first. If viewers reach it and few click, read the button label and destination together. To give each prospect a personal link and see how much each one watched, see personalized video emails at scale.

Choose what to change when the numbers are low

Start with the earliest point where the viewer’s path breaks down. A video with few views may need a clearer link or a more specific reason to open it. A video with plenty of views and an early drop-off gives you a place to look in the recording itself.

What you seeWhat to checkWhat to try
Few views and few unique viewersWho received the link and where it appearsPut the link by a sentence that says what the viewer will learn
Views grow and unique viewers barely moveWhether the same audience is returningShare the video with another relevant group if you need more reach
Low watch percentage and an early dropHow soon the promised step appearsOpen on the task and explain the setup as you go
A cool section before a bright one on ProWhich part answers the reader’s questionMove the sought-after step earlier and trim the lead-in
Viewers reach the demo and few click the CTAThe label and the page it opensGive the button one action with a clear destination

Pick the row that matches your goal and change one thing for the next send. Write down what you changed and the point you expect more viewers to reach. When you check again, use a similar time window and look at where people came from so you can put any change in context.

Frequently Asked Questions

Which video metric should I check first? Start with unique viewers if you want to measure reach. For a demo, find the timestamp where the viewer sees the result and check retention there. If you ask viewers to take a next step, such as booking a call, check CTA clicks if you used a button.

How do I compare engagement for videos of different lengths? Start with average watch percentage, then look at the timestamp of the step you want people to see. A long training video and a short sales demo give viewers different jobs. Compare videos with a similar purpose and audience, and write down where the main task appears in each recording.

What if people leave just after they see the answer? Think about what they came to do. If a support video shows where to find a setting, leaving after that step may mean they got what they needed. Watch the section after the drop and decide whether it contains a step the viewer still needs to complete the task.

Why does a video shared by email show Direct traffic? The app that opens the link can omit its source, which makes the visit appear as Direct. For the next email, add utm_medium=email to the share link so it appears as Email. Keep track of when you switch to tagged links so you can compare the two periods with that change in mind.

What if I have only a few video views? Use the report to find specific moments worth checking and collect more views before drawing a broad conclusion. Play the video from a prospect’s point of view, check the first step on a phone, and make sure the link is easy to find. Then share it with another relevant group and review the same metrics.

Ready to transform your video workflow?