To measure push notification performance, track five core metrics: opt-in rate, delivered count, click-through rate (CTR), unsubscribe rate and conversions from push traffic. Add subscriber growth and expired subscriptions to watch list health. Measure CTR on delivered notifications, compare campaigns against your own history, and tag every link with UTM parameters for analytics.
Most push dashboards show plenty of numbers. The trick is knowing which ones change your decisions. When I review a push program, I start with the list (is it growing and healthy?), then the campaigns (are people clicking?), then the business result (did those clicks do anything?). Below are the metrics that matter at each step, with formulas, and how to read them without relying on someone else’s benchmark.
Which push notification metrics should you track?
Track metrics at three levels: list health, campaign engagement and business outcomes. This table is the set I use.
| Metric | Formula | What it tells you |
|---|---|---|
| Opt-in rate | New subscribers ÷ visitors shown the prompt | How well your prompt and timing work |
| Subscriber growth | New subscribers minus unsubscribes and expired, per week | Whether the list is growing |
| Delivered | Notifications accepted and shown by browsers | How many people could see the campaign |
| Click-through rate (CTR) | Clicks ÷ delivered × 100 | How interesting the message was |
| Unsubscribe rate | Unsubscribes ÷ subscribers in the period | Whether you’re sending too often or off-topic |
| Expired subscriptions | Endpoints returning 404 or 410 | Natural list decay |
| Conversions | Goals completed from push sessions | Business value of the channel |
| Revenue per 1,000 delivered | Push revenue ÷ delivered × 1,000 | Value per send, for comparing campaigns |
You don’t need all eight on day one. If you track only three, make them opt-in rate, CTR and conversions.
How do you calculate push notification CTR?
Push notification CTR is clicks divided by delivered notifications, multiplied by 100. If a campaign is delivered to 20,000 subscribers and gets 600 clicks, the CTR is 3%. That’s an example to show the maths, not a benchmark.
Use delivered rather than sent as the base where your tool reports it. “Sent” includes messages to expired endpoints and offline devices, so it makes CTR look lower than it is. Also, compare CTR across campaigns of the same type. A breaking-news alert and a weekly digest behave differently, so mixing them hides what works.
What is the difference between sent, delivered and clicked?
Sent is what your server handed to the push services, delivered is what reached browsers, and clicked is what people tapped. The gaps between them point to different problems.
- Sent but not delivered: expired subscriptions, devices offline past the message’s time to live (TTL), or push service errors. Clean the list and check error codes.
- Delivered but not clicked: weak title, wrong timing or wrong audience. Work on copy, timing and segments.
- Clicked but no conversion: the landing page doesn’t match the promise. Fix the page, not the notification.
If delivered drops suddenly for everyone, treat it as a technical issue first. My checklist on why web push notifications are not showing covers the usual causes.
Can you track open rates for web push?
Not reliably. Web push has no open event like email, because the notification appears on screen and the user either clicks, dismisses or ignores it. Some tools report “impressions” or “displayed”, but clicks are the dependable engagement signal. I explain this further in whether you can track open rates for push notifications.
How do you measure conversions from push notifications?
You measure push conversions by tagging every notification link with UTM parameters and reading the results in your analytics tool. For example, add utm_source=push, utm_medium=web-push and utm_campaign=your-campaign-name to the click URL.
Then, in Google Analytics 4, push traffic appears under that source and medium, and you can attach key events such as signups or purchases. Google’s URL builder and UTM guide explains the parameters. For step-by-step setup, see how to track push notification visits in Google Analytics.
How should you read push metrics without benchmarks?
Compare against your own history, not a published average. Industry CTR and opt-in figures vary widely by niche, audience, prompt type and sending frequency, so a number from another site tells you little about yours.
Here’s the approach I recommend:
- Set a baseline. Record your median CTR and opt-in rate over the last 30 days.
- Change one thing at a time. Test a new prompt, title style or send time, and keep everything else fixed.
- Read medians, not single campaigns. One viral story can skew an average.
- Watch unsubscribes with CTR. More sends may raise total clicks while quietly raising unsubscribes.
- Segment the view. Break results down by device, browser and country before drawing conclusions.
For timing tests, my post on the best time to send push notifications is a good place to start, and push notification frequency covers how often to send.
Which reports should you check weekly vs monthly?
Check campaign metrics weekly and list health monthly. That rhythm catches problems early without drowning you in data.
| Rhythm | Check | Action if it slips |
|---|---|---|
| After each campaign | CTR, clicks per CTA button | Adjust title, image or audience next time |
| Weekly | Median CTR, unsubscribes, new subscribers | Revisit frequency and prompt timing |
| Monthly | Subscriber growth, expired subscriptions, conversions | Review the prompt, segments and landing pages |
| Quarterly | Revenue or goals per 1,000 delivered | Decide where push sits in your channel mix |
Where does LaraPush fit?
LaraPush is a self-hosted push panel whose analytics show sent, delivered and clicked counts per campaign, with CTR and breakdowns by date, location, device, OS and browser. It also tracks clicks for each CTA button separately, and Pro adds advanced analytics. Because it runs on your server, the subscriber data stays with you. Trade-offs: you run a VPS, a licence covers one server, and there’s no free plan. See features, audience segmentation and pricing. To compare tools, see the best push notification services and tools.
Frequently asked questions
Track opt-in rate, delivered count, click-through rate, unsubscribe rate and conversions from push traffic. Add subscriber growth and expired subscriptions to monitor list health. If you only track three, choose opt-in rate, CTR and conversions, because they cover the prompt, the message and the business result.
Divide clicks by delivered notifications and multiply by 100. For example, 600 clicks from 20,000 delivered notifications is a 3% CTR. Use delivered rather than sent where possible, because sent includes expired subscriptions and offline devices that could never click.
It depends on your niche, audience, prompt and frequency, so published averages are a weak guide. A better approach is to set a 30-day baseline for your own median CTR, then test one change at a time and keep what beats your baseline.
Add UTM parameters to every notification link, such as utm_source=push and utm_medium=web-push, plus a campaign name. In Google Analytics 4, filter by that source and medium and mark signups or purchases as key events to see conversions from push.
Some sends go to expired subscriptions, which return 404 or 410 errors, and some devices stay offline until the message’s time to live runs out. Remove expired subscriptions regularly and use a sensible TTL so time-sensitive messages don’t arrive too late.
