Why it isn't converting
Why is my conversion rate so low: split the rate
The blended number hides the visitors who do sign up, and it hides sign-ups your tracking misses. Mimiq sends simulated visitors through the step that loses the most, and shows where they stop.
Why is my conversion rate so low?
A conversion rate is completed actions divided by visits in one group, so a low number has three possible causes: a mix of visitors who were never going to buy, sign-ups that the tracking misses, or a real break in the path. Split the rate by traffic source, device and landing page before you change anything. Then check the sign-ups against your own database. Mimiq covers the last cause: it sends simulated visitors down the path and reports the step where each one stopped and why.
What this looks like in a report
“I have never sent a reminder with this. The plans page made me pick a tier before I knew what any of them does. I will try the one I already know and come back to this later.”
The fix. Put a working reminder example above the plans, with a button labeled Send my first reminder. Move the tier choice to after the first reminder is sent.
An illustration. Run a test to see your own.See a real sample report
One average hides the groups and the missing sign-ups
A conversion rate is a ratio, and a ratio can be low for reasons on either side of the division. On the bottom, the visits include people who came to read, bots, your own team and returning logins. A broad social post can send many visits from people with no reason to sign up today, which pulls the blended number down even when a search visitor who wanted your product signs up at a healthy rate. On the top, the sign-up event may miss sign-ups entirely, so a page that works still reports a low rate.
- Sources: which channel sends most of the visits and few of the sign-ups?
- Devices: is the rate on phones far below laptops for the same page?
- Landing pages: do the busiest pages share a headline the buyers never see?
- The event: does one sign-up create one event, or does a thank-you refresh fire it again?
Split the rate before you judge the page
Pick one conversion event and one date range, then split the rate three ways: by traffic source, by device and by landing page. A blended number can be low because one channel sends many visits and almost no buyers, while search visitors convert at a healthy rate. For example, 2 sign-ups from 40 visits from a broad social post and 9 sign-ups from 30 visits from a search for your exact product are two different groups. Judging the page by the blend would send you to rewrite a headline the search visitors never had a problem with. A low rate that shows up on phones only is a layout problem, and the split tells you which one you have.
Check the sign-ups against your own database
Analytics can miss sign-ups. An ad blocker or a privacy setting can stop the tag from loading, and an event that fires only on one page load misses anyone who closes the tab before it runs. Count the new accounts in your product for the same days, and compare that number with the sign-up events. If the database is higher, the rate is low because the count is short, and the page may be fine. The opposite error happens too: an event that fires again when the thank-you page is refreshed inflates the count.
Test the segment that loses the most visitors
Once the split shows a segment that is losing people, run the test as a visitor from that segment. Set the arrival to match it: a broad post, or a search for the category rather than your product name. Then put the visitor's reason in the task, such as comparing reminder tools for a small office. The report shows where that visitor stopped, what they expected to see there and the fix to try first.
Four ways to see the path, and what each one misses
| Method | What it shows you | What it cannot show you |
|---|---|---|
| Analytics funnel | How many visits reach each step, and how many finish it | Why a step loses people, or whether the tag saw every sign-up |
| Session replay | What real visitors clicked and scrolled on a page | Visitors who never reached the page, or who blocked the recording script |
| Scripted end to end test | Whether a known path still works on each new build | What a stranger expects, because the script already knows the answers |
| Simulated visitors | Where a visitor with a reason to come stops, and what they thought | Real conversion rates, or how many real people will sign up |
Use the first two for rates on real traffic. Use the last two to find the broken step before you send more people to it.
What you get with Mimiq
- 01
Run it by hand, or on every pull request
Paste a link and a task in the app and watch the visitors work. Or run them on every preview deploy: the GitHub Action comments on the pull request, any other CI can call the MCP endpoint, and your coding agent can run it too.
- 02
Bugs and usability problems in one report
The browser records errors and failed or slow requests on the step they happened. The visitors say where the wording or the steps confused them, and where they left.
- 03
Visitors that behave like your audience
Each visitor is a different profile with a reason to be there, matched to who the page is for, described by you, or built from your own customers.
- 04
A prompt to fix each problem
Every problem comes with a fix to paste into Cursor, Claude Code, Lovable or any coding agent, and one prompt hands over the top problems at once. Then send the same visitors through again.
- 05
Follow every visitor
See each visitor's whole path, step by step with screenshots, and what it was thinking at every step. Ask any visitor a question.
The browser records relevant JavaScript errors and failed or slow app requests encountered during the run, attached to the step where they occurred. The visitors are simulated: they find problems, they do not predict how many real people will convert.
A new broad post can drop the blended rate without any page change
Each new channel changes the mix of visitors, so the blended rate can fall with no change to the page. After you change a page for one segment, rerun a simulated visitor from that segment through to the sign-up, to see whether they reach the form with the same reason to stay. Keep the test on the segment that loses the most people, and check the event count against the sign-up count again after every change to the tag.
Starter
Try it on one launch
About 4 tests of 5 visitors
- Visitors on your landing page, signup and checkout
- Where they left and why, with a fix to copy for each problem
- Run it in the app or on every pull request
- Ask any visitor a question, included
- Test accounts and protected previews
Builder
Test every build
About 20 tests of 5 visitors
- Everything in Starter
- Credits refill each month, and what is left rolls over once
- Cancel anytime
Pro
Test every release, as a team
About 60 tests of 5 visitors
- Everything in Starter
- Credits refill each month, and what is left rolls over once
- Cancel anytime
Scale
Test at scale, every day
About 240 tests of 5 visitors
- Everything in Starter
- Credits refill each month, and what is left rolls over once
- Priority support
- Cancel anytime
Questions
What is a good conversion rate for my site?
The number depends on the offer, the price and where the visitors came from. Compare your segments with each other and with your own past numbers. A simulated test does not give you a rate. It does not predict how many real visitors will sign up, and no simulated run forecasts a conversion rate for your site.
Can a test tell me why my visitors do not sign up?
It can show where a simulated visitor with a reason to come stops, and what they were thinking at that step. That is a reason to check, not proof of what happens in your traffic. Pair it with the segment split. If the simulated visitor stops at the same step your weakest source stops at, you have a real lead to follow.
My analytics shows more sign-ups than my database. Which one is right?
For accounts, the database is the one to trust, because it holds the accounts that exist. Analytics misses sign-ups when a browser blocks the tag or when the event fires on one page load only. Check both for the same days, and look at the days where they split most.
Should I test every traffic source?
No. Start with the source that has the most visits and the lowest rate, because it pulls the blended number down the most. A source with a handful of visits does not tell you enough to change a page, so wait for more visits before you act on it.
Do simulated visitors count toward my conversions?
They can. If your sign-up works and the visitor can get the confirmation email, a simulated visitor can finish it, and that creates a real account in your product. Delete the test accounts after each run, or they stay in your sign-up count and make the rate look better than real traffic does.
Keep reading
Send visitors and see what breaks first.
Your first test is free: sign in with Google and it starts right away. Last checked 2026-11-02.