Website analytics · Beta
See where visitors leave – and what you can do about it.
Lumeya shows you the spot on your website where visitors drop off. You get the reason in plain words and a suggestion for what to change.
Try it
- 1Click the section marked in red, “Prices”.
- 2Let Lumeya rework it, or hide the tip for a while.
- 3Switch between version A and B at the top.
Every number in this demo is sample data from a made-up hair salon.
Analytics
Salon MiraBetaSample dataLast 30 days
1 spot needs your attention
Lots of visitors leave here. The tips show you what you can change.
1 important1 to check0 ignored
- Page views
- 1,235
- Pages served, not people
- Leads per 100
- 2.3
- 28 leads in total
- Mobile share
- 68%
- Most people come on a phone
Your sections
Version A · A is your page as it is live right now.
Bar: how many visitors get this far. Striped: how many leave here.
Ignored tips0
Tips you hide for a while end up here. You can bring them back any time.
Tips for you
Click a tip or a section to see more.
Split test: version A and B
At the prices, fewer visitors leave in version B. For the page as a whole, the data is too thin for a clear verdict.
| Section | A | B | Result |
|---|---|---|---|
| Whole pageEnquiriesno clear difference | 11 of 620 | 17 of 615 | no clear difference |
| PricesDrop-offsweaker in A | 41% (434) | 29% (431) | weaker in A |
| ContactDrop-offsno clear difference | 31% (198) | 31% (240) | no clear difference |
Measured page views, not people. A difference does not prove the cause.
Demo with sample data. Nothing is saved or sent.
Counts without cookies
Lumeya counts page views, scroll depth and clicks as daily totals. No cookie, no visitor ID, no third-party trackers.
Whether visitors see a notice or give consent first depends on the country. Lumeya picks the matching rule.
Maths first, then AI
First a fixed algorithm checks your numbers. An AI review only runs when something really changes or when you start it.
You see what a review costs before it starts.
Draft first, then live
Lumeya makes every change as a draft. You publish it yourself or test it as version B against your current page.
Every forecast is an estimate, not a promise.