By Fadi Agour, J.D., Founder & CEO | Last updated October 7, 2026
Fake traffic shows up in the clock before it shows up in the lead. Time-to-convert is the time between a click and the conversion that follows it. Measure it for every lead or call, by source and sub ID, and fake traffic stands out in the shape of the data.
Key takeaways
- Time-to-convert is the gap between a click (or landing) and a form submit, call start or connected call.
- Judge the shape of many conversions, not one.
- Four shapes deserve a closer look: a pile-up at the fast end, a flat curve, a tight repeated interval, and a sudden change from last week.
- A timing flag means “look closer.” Two markers together mean “hold payment until a person has checked.”
- Set your thresholds from your own baseline, never from a number in someone else’s article.
What is time-to-convert?
Time-to-convert is the elapsed time from a click (or landing) to the conversion: a form submit, a call start, or a connected call. Some teams call it time to conversion, time to action, or click-to-conversion time.
One number tells you little. The distribution of many numbers tells you a lot.
Why real people leave a recognizable pattern
A real person lands on a page, reads, hesitates, scrolls, then picks up a phone or types a name. Some are quick. Most take a normal amount of time. A few take much longer, because they left and came back.
That gives you a spread: a few fast ones, a bulk in the middle, and a tail. The spread is messy, and the mess is the proof.
Software does not hesitate. It acts on a schedule, on a script, or with no schedule at all. That leaves a different shape.
Four shapes that should make you look closer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Any one of these is a reason to look. None of them is proof.
How to set up time-to-convert tracking
- Log the click time and the conversion time for every lead or call. Tie them together with a click ID, not a cookie. (Our post on server-side tracking vs pixels explains why.)
- Group by source and sub ID. A clean publisher can hide one bad sub ID. A bad average can hide a clean one.
- Build your own baseline. Use the first clean weeks of a campaign. A legal form, a home services call and an insurance quote do not convert at the same speed.
- Set the “too fast” line from your baseline, and review it every month.
- Compare week over week. A healthy source has a stable curve. A change is the signal.
When nobody converts, watch the steps
Time-to-convert needs conversions. Sometimes a source sends none.
In late September we tested a new paid channel on one of our legal landing pages. Our own log, for four days, showed 184 paid clicks and 157 landings. Only 13 visitors tapped the first question, which takes one tap. Three answered the second. None submitted. More than 90 percent of the visitors who reached the form never touched it.
That did not prove the clicks were fake. People click out of curiosity, and a topic match is not the same as intent to hire a lawyer. So we paused spend and asked the platform to check the signals only it can see. The funnel steps told us where to look.
Two markers, not one
We do not cancel on one flag. A timing flag alone means “look closer.”
Pair it with a second marker and it becomes a hold: repeat callers above normal, a geo that does not match the order, calls that all end just past the minimum duration, a burst outside business hours, or a sudden surge.
Two markers together stop payment review until a person has listened to calls or read submissions. That protects honest publishers from being cut off after one bad day, and it protects the buyer from paying for noise.
Pause the smallest thing that is bad
Do not shut down a whole publisher for one bad sub ID. Hold payment on that sub ID, sample at least ten calls or submissions, and write down what you found. Release it only after a clean sample and a written answer on what changed.
That written record does a second job. If a buyer disputes a lead later, it is your evidence.
What time-to-convert will not tell you
It does not prove fraud. A fast conversion can be a returning customer. A slow one can be a person who called back after lunch.
It also does not replace listening. Pay-per-call buyers still need call duration, repeat-caller and recording review. Pay-per-lead buyers still need form review and verification. Time-to-convert tells you where to point those tools.
Where this fits at RPM
We have run performance marketing since 2015, and RPM has made the Inc. 5000 seven years in a row. Our rule is simple: volume without quality isn’t a lead business. It’s a noise business. Timing checks are one of the ways we keep the noise out of our buyers’ pipelines.
Frequently asked questions
What is time-to-convert?
It is the time between a click (or landing) and the conversion that follows it: a form submit, a call start, or a connected call. The pattern across many conversions is the test, not any single value.
What is a normal time to conversion?
There is no universal number. It depends on the vertical, the page and the traffic source. Build a baseline from your own first clean weeks and compare each source against it.
Can time-to-convert detect bots on its own?
No. It flags suspicious patterns. Pair a timing flag with a second marker, such as repeat callers or a geo mismatch, then review a sample of calls or submissions.
Does time-to-convert work for pay-per-call and pay-per-lead?
Yes. For calls, measure click to call start, then add call duration and repeat-caller checks. For leads, measure click to form submit.
How often should I review time-to-convert?
Daily for new sources. Weekly for established ones, with a week-over-week comparison of each source’s curve.
Part of the RPM Insights library.