Product analytics Understand every user. Find your next growth lever.
Behavioral analytics, user profiles, marketing attribution and growth dashboards — from "how many came" to "who they are, what they did and why they stayed." No tracking plan, no data team required.
- Autocapture + AI event names
- Web, app and mini-program as one user
- Free plan available
Ask a question
A scripted demo — no real AI is called on this page.
Trial → paid funnel · last 30 days
Signed up
6,120
Created a project
4,406
Installed snippet
2,203
Paid
612
Half of new projects never install the snippet. Users who install within 24 hours pay 4.1× more often. Send a setup nudge on day one.
Weekly retention · new sign-ups
| Cohort | Users | W0 | W1 | W2 | W3 | W4 |
|---|---|---|---|---|---|---|
| Sep 1 | 1,840 | 100% | 46% | 38% | 33% | 31% |
| Sep 8 | 2,012 | 100% | 49% | 41% | 36% | |
| Sep 15 | 1,966 | 100% | 52% | 44% | ||
| Sep 22 | 2,130 | 100% | 58% |
Week-1 retention rose from 46% to 58%. The lift started with the Sep 22 cohort — the week the onboarding checklist shipped.
Revenue ÷ ad spend by channel · last 30 days
Newsletter returns $6.20 per $1. TikTok spends 23% of the budget for 7% of orders — the first place to cut.
Product analytics answers three questions
What did users do?
Behavioral analytics
- Events and formulas
- Funnels and drop-off
- Retention and paths
- Click and scroll heatmaps
Who are they?
User analytics
- One profile across devices
- Properties and distributions
- Rule-based and RFM segments
- B2B account view
Where does growth come from?
Marketing & growth analytics
- Ad click IDs and UTM
- Channel ROI with imported costs
- AARRR dashboard
- Revenue, AOV and LTV
See every step users take
Six models built on the same events — switch views without rebuilding anything.
Funnels
Find the step where people drop, and who they are.
Product view
10,000
Add to cart
4,200
Checkout
2,270
Paid
1,630
Events
Multiple metrics and formulas on one chart.
Retention
Did new users come back?
| Cohort | Users | W0 | W1 | W2 | W3 |
|---|---|---|---|---|---|
| Sep 8 | 1,204 | 100% | 44% | 36% | 31% |
| Sep 15 | 1,318 | 100% | 47% | 39% | |
| Sep 22 | 1,276 | 100% | 51% |
Paths
What people do after the pricing page.
Distribution
Orders per buyer, last 90 days.
Heatmaps
Where people click and how far they scroll.
Scroll depth
100% → 22%
Sample data
Hand the tracking plan to AI
Autocapture records clicks, submits and scrolls from day one. AI groups them into candidate events and names them; you confirm in the event inbox. When an event needs business fields, AI writes the snippet.
Autocapture
Clicks, form submits and scrolls are recorded as soon as the snippet is live.
AI suggests events
Similar interactions are grouped and given readable names.
Confirm in the inbox
One click turns a candidate into a reportable event.
Add business fields
For orders and plans, AI generates a ready-to-paste snippet.
Captured interactions · today
- click · button.cta "Start trial"1,204×
- submit · form#signup642×
- click · a[href="/pricing"]2,318×
- scroll · /pricing · 75%980×
AI suggestions
- trial_started96%
Clicks on "Start trial" on any page
- signup_submitted94%
Sign-up form submitted
- pricing_viewed81%
Pricing page reached 75% scroll
Event inbox
- trial_startedConfirmed
- signup_submittedConfirmed
- pricing_viewedConfirm
Generated snippet
// Paste after payment succeeds
tapcub.track('order_paid', {
order_id: order.id,
amount: order.total,
currency: 'USD',
});Traditional tracking
Weeks
Write a plan → schedule → develop → verify. Every new question waits for a release.
TapCub
Minutes
Install and collect. Confirm the events AI suggests and they are ready to report.
From visitor counts to real people
Profiles, identity, properties and segments — so you can see who converts, who churns and who to talk to.
- Profiles — one timeline across web, mini-program and app
- Identity — devices and pre-login visits merged automatically
- Segments — rule-based or RFM, updated continuously
- Accounts — roll users up to companies for B2B
Lena M.
Berlin · 3 devices · since Jun 2026
14
Sessions
$412
LTV
3
Orders
- Read /blog/linen-care on the websiteSep 3 · Google
- Ordered in the mini-programSep 9 · shared link
- Repeat purchase in the appSep 28 · push
Segment · High-value, at risk
Matching users
1,284
RFM · recency × frequency
Account for every advertising dollar
Ad click IDs, UTM, first-touch attribution and landing-page conversion. Import costs and see channel ROI on one quadrant.
- Click IDs from 14 ad platforms recognized on arrival
- First-touch and last-touch attribution
- Landing-page conversion by campaign
- Cost import for ROI and ROAS
Ad click IDs recognized automatically
Channels · spend share × ROI
Newsletter returns 6.2× on a small budget; TikTok takes a large share at 0.8×.
Keep growth on one dashboard
AARRR stages, goals, revenue and a product health score — with industry templates to start from.
A · Acquisition
48,213
Visitors
B · Activation
6,120
Sign-ups
C · Retention
42%
Week-1 retention
D · Revenue
$86,400
GMV
E · Referral
8.4%
Invited a friend
Product health score
Activation and retention up; referral flat for 3 weeks.
Revenue
GMV
$86,400
Orders
1,168
AOV
$74
LTV
$212
Goals and conversion
Track sign-ups, purchases or any event as a goal, with conversion rate over time.
Revenue analysis
GMV, AOV, repeat purchase and LTV by channel and segment.
10 industry templates
Start with the metrics and dashboards that fit your business.
Let the data tell you first
Ask in plain language, get explanations for changes, and be alerted when something moves.
Ask in plain language
Get a chart and a one-line conclusion.
AI insights
Notable changes explained by segment and channel.
Anomaly alerts
Spikes and drops sent where your team works.
Subscriptions
Daily or weekly digests of the charts you pin.
Recent alerts
- Sign-ups · mobile −38%Tue 09:40 · submit error rate 2% → 41%WeCom
- Checkout exit rate above 45%Mon 08:00 · 2 days in a rowEmail
- ChatGPT referrals +41%Sun 07:00 · week over weekFeishu
Delivered by email, WeCom, DingTalk, Feishu or webhook.
Sample dataWebsite, app and mini-program count as one user
One project takes several data sources. Identity resolution joins them, so funnels and retention follow the person, not the device.
- Multiple sources in one project
- One user definition across platforms
- Platforms side by side in every report
Web
example.com · /pricing
Mini-program
WeChat · pages/order
App
iOS · v4.2
Lena M.
One person · 3 devices · 14 sessions
Clear for beginners, deep for experts
Switch any time. Both modes read the same data.
Five entries, charts on arrival
Pick a question and see the answer — no model to configure.
Full models, formulas and SQL
Every analysis model, custom metrics and formulas, and SQL when you need it.
-- Buyers by channel
SELECT channel, count(DISTINCT person_id) AS buyers
FROM events WHERE event = 'order_paid'
GROUP BY channel ORDER BY buyers DESC;Built for teams and for your stack
Lighter than traditional analytics platforms
| TapCub | Overseas product analytics | Enterprise analytics platforms | |
|---|---|---|---|
| Time to first chart | Minutes | Days | Weeks |
| Tracking plan required | No — autocapture + AI | Usually | Yes, with implementation |
| Data team needed | No | Often | Usually |
| Website analytics included | Yes | Separate tool | Add-on |
| Live chat included | Yes | No | No |
| Free plan | Yes | Limited | No |
| Self-hosting | Enterprise | Some | Yes |
Category comparison based on public product information. Individual vendors differ.
Start with a template for your industry
Ten templates come with the metrics, funnels and dashboards that matter for each business.
Questions about product analytics
Still have a question?
Chat with the team behind TapCub — we usually reply within a few hours on business days.
How is this different from website analytics?
Website analytics answers "how much traffic and from where". Product analytics follows people: events, funnels, retention, profiles and segments. Both run on the same snippet and the same data.
Do developers have to write tracking code?
Not to start. Autocapture records clicks, submits and scrolls, and AI suggests named events. You only add code for events that need business fields such as order amount — and AI writes that snippet for you.
Can it be self-hosted?
Yes. TapCub Enterprise can be deployed on your servers or private cloud and is quoted separately. See Enterprise.
Does it slow down with a lot of data?
Reports read from a columnar store built for event data. Paid plans raise event volume; Enterprise contracts set limits per agreement.
Can I export to my own data warehouse?
Any report can be exported as CSV, and the open API returns metrics and events. Webhooks push events to other systems in real time.
Is product analytics a separate subscription?
No. Product analytics is part of the analytics plans; some features such as autocapture and identity resolution start on Pro. See pricing.
Need self-hosting?
Keep your data on your own servers. TapCub Enterprise is quoted separately for your security and compliance needs.