How Do You Know If Your Website Is Bringing You Phone Calls?
How I matched a law firm's website phone taps to its real calls, the clock trap that hid every match, and the steps to check your own this month.
A tap on your website's phone number only proves a finger touched a link. Your phone system's call history is its list of calls and their times. Compare each tap's time with that list, and count a match when an incoming call followed within a few minutes. Then check how often a call follows a random moment that fast, so you know what luck alone looks like. For a law firm I work with, 18 of 20 October taps were followed by a call within three minutes, against about 3 you'd expect by chance.
Key takeaways
- A phone tap opens the phone's calling screen. Check the call history to see whether an incoming call followed.
- Write down both systems' time zones before you compare anything.
- Check whether calls followed taps more often than they followed randomly chosen times.
- Keep each call's time, length, direction and whether it was answered, never callers' numbers or names.
Why isn't a tap the same as a call?
A phone tap is someone touching the phone number on their phone screen, which opens the phone's calling screen. A website tap and an incoming phone call are two different records, kept in two different places. I build the monthly marketing report for a law firm that gets most of its new clients by phone, and one month showed 20 phone taps and 38 contact forms. That made the taps about a third of all the leads in the report, so the question was whether those taps were followed by incoming calls.
What information did I keep?
I asked the firm before connecting its phone records to the report. The email said I would keep only each call's time, length, direction (incoming or outgoing) and whether someone answered, never callers' numbers or names. The firm said yes.
The connection has one permission: read the list of calls. It drops everything except those four details before saving anything, so the number of phone numbers kept is zero. That was all the information I needed to compare incoming calls with taps on the website.
Why did nothing match at first?
The two lists used different clocks. Google Analytics recorded the taps using its reporting time zone, so I needed to check which time zone that was. The phone system reports in universal time, so my first comparison found no matches at all.
Instead of guessing, I tested 17 possible time differences, from eight hours behind to eight hours ahead, and counted the matches for each. A seven-hour difference produced far more matches than the others. Analytics was on Pacific time, and accounting for that difference showed which taps were followed by incoming calls.
You can look up your own reporting time zone in Google Analytics under Admin, then Property details. Google's help page shows where to find the setting. Write that time zone down next to your phone system's time zone before you compare a single call.
How do you know the matches aren't luck?
The firm takes about twenty calls a day, so a call can follow a tap by chance. I checked 4,000 random weekday moments during office hours and counted how often a call followed within three minutes. It happened about 16 percent of the time, roughly 1 in 6.
That's the chance rate: how often a call follows a moment when nobody tapped anything. In my experience, comparing the two is what shows whether calls followed taps more often than they followed random moments. On a busy phone line, timing alone can look far more convincing than it is.
What did the numbers show?
- September 2026: 10 of 10 phone taps had an incoming call within three minutes, against about 1.6 expected by chance.
- October 2026: 18 of 20 had an incoming call within three minutes, against about 3.2 expected by chance.
Those counts sit far above the chance level, so the evidence says most taps were real people calling. They still don't prove which caller tapped which number. My first chart headline said "Taps are real calls," and a second AI reading the report as the client would flagged that as too strong, so it now reads "Most taps turned into calls."
The report also gave the staff a number behind a request they had already made, to keep the phones open later in the afternoon. The main line now rings until 4:30. The comparison runs every month, saves counts only and adds one line under the phone taps chart.
An automatic check warns me if calls start following taps about as often as they follow random moments, which would mean the comparison is no longer working. It's the same habit I wrote about in how to know if your backup actually ran. Check the record of what happened instead of stopping at the number a tool reports.
How do you check your own phone taps this month?
You need last month's tap times from your website report and last month's incoming call times from your phone system:
- Ask whoever owns the phone account whether you can download a call history showing each call's time. Agree to keep only the time, length, whether the call came in or went out and whether it was answered, never callers' numbers or names.
- Look up your Google Analytics time zone under Admin, then Property details, and write it beside the phone system's time zone.
- List every tap's time and every incoming call's time, using the same time zone for both lists.
- Count a tap as a call when an incoming call came within a few minutes. This firm's report uses three.
- Pick a handful of random times during office hours and check how often an incoming call followed within the same few minutes you used for the taps.
- If calls followed taps much more often than they followed random times, describe the result plainly, such as "Most taps turned into calls."
Use the hours your phone line actually rings when you pick those random times. My first version assumed weekdays from 9 to 4, but this line also rings Saturday mornings and, from November, until 4:30 on weekdays. The report now follows the phone system's real schedule.
Frequently asked questions
Do I need permission to use the call records?
Ask whoever owns the phone account before you use its records in a report. Say exactly what you'll keep and what you'll leave out. For this firm I asked in a short email and got a yes before connecting anything.
What about calls that went to voicemail?
This phone system's records don't show voicemail. So the report says "not answered," because that's what the records establish, and never calls those calls "lost."
How close does a call have to be to count?
I used three minutes for this report. Also check how often an incoming call followed a randomly chosen time during office hours within those same three minutes. If calls follow random times almost as often as they follow taps, that time limit doesn't give you strong evidence that the taps became calls.
Can I do this without a developer?
Yes, with the two downloaded lists, as long as both include the times. Put both lists in the same time zone first, then check each tap against the incoming calls. A script, which is a small program that repeats the same steps, can make the comparison automatic every month.
Help checking the numbers in your marketing report
I build monthly marketing reports for clients as part of my small business marketing work. If you'd like help with yours, tell me what your website counts as a phone lead. Start this week by asking whoever manages your phone account whether you can download last month's incoming call times.
Want a system built for your business?
Everything here comes out of real work, mine and my clients'. If you want something custom, let's talk.
Get in touch