Why Your Call Tracking Numbers Are Probably Lying To You

Almost every contractor I talk to thinks they answer most of their phone calls. They will say things like 'we get the phone on the first or second ring' or 'my dispatcher is great, she catches everything.' Then we pull their call logs and the actual answered-call rate is somewhere between 40% and 65%. The gap between perception and reality is enormous, and it persists for a specific reason: most contractors are reading the wrong metric. The metric most phone providers default to showing is 'calls answered', but the right metric is 'calls answered as a percentage of all inbound attempts.' Those are different numbers.

Here is the most common version of the lie. A contractor's call tracking dashboard shows 'Calls Answered: 240 this month.' That number feels good. What it does not show is the 180 calls that were attempted but never connected: calls that hit a busy signal because two lines were already in use, calls that rang out because no one picked up, calls that were forwarded to voicemail because the dispatcher was on another call, and calls that hit an after-hours message. Those calls do not appear in the 'answered' count, but they also might not appear prominently in the 'missed' count if your dashboard is showing the wrong default view.

The number to look at is 'inbound call attempts' versus 'connected calls.' Some dashboards label these as 'total inbound' versus 'completed.' Others use 'sessions' versus 'answered.' The naming varies. What you are looking for is the ratio: out of every inbound attempt, what percentage actually resulted in a human (or AI) conversation. Industry research from HomeAdvisor and similar sources consistently puts this number at 38% for home service contractors, meaning 62% of inbound attempts never result in a real conversation. That is the number your dashboard is probably hiding.

The second source of the lie is voicemail. A voicemail-answered call typically shows up in the 'answered' column on most dashboards. That is technically accurate. The system did answer, just with a recording. But voicemail is not a useful answered state for a service business. Google and Ipsos have measured that roughly 80% of consumers will not leave a voicemail when calling a local business; they will hang up and try the next listing. So 'calls answered' that ended in voicemail are functionally the same as 'calls missed' for business purposes. If your dashboard does not separate human-answered from voicemail-answered, you cannot tell the difference.

The third lie is hold time. A call that connected but sat on hold for four minutes before the dispatcher finally got to it is technically 'answered.' But on the customer side, 78% buy from the first responder (HBR 2011), and 'first responder' means whoever picks up promptly, not whoever picks up eventually. A four-minute hold gives your competitor a generous head start. The right metric is 'calls answered within 20 seconds' or 'calls connected without hold.' Most dashboards bury this.

Pull your phone provider's most detailed call log for the last 30 days and run this audit. Total inbound attempts. Calls that connected to a human. Calls that hit voicemail. Calls that rang out. Calls that hit a busy signal. Average time-to-connection for the calls that did connect. Calculate the ratio of human-connected to total attempts. If you are above 80%, you are doing exceptionally well. Between 60% and 80%, you have room to improve. Below 60%, the missed-call problem is materially hurting your revenue and you should treat it as a top-three operational priority.

The math from there is straightforward. Suppose your audit reveals 100 inbound attempts per week, 60 connected, 40 missed or voicemail. Use a conservative average ticket of $280 (HomeAdvisor) and a conservative conversion rate of 40%. The missed-call revenue exposure is roughly 40 × 0.4 × $280 = $4,480 per week, or about $19,000 per month. Even if your numbers are half of that (20 missed calls per week, 30% conversion, smaller average ticket) you are still talking about thousands of dollars per month that your dashboard is not surfacing because it shows you 'calls answered' instead of 'calls attempted versus connected.'

The fix is not 'try harder to answer the phone.' The fix is recognizing that during peak hours, lunch breaks, after hours, and during multi-call surges, no human-only setup can answer everything. The structural fix is some form of overflow capacity, most commonly an AI voice agent that can handle simultaneous calls without queueing, paired with a measurement system that shows you the real numbers. Once you can see the real numbers, the operational decisions become obvious.

Run the audit this week. Spend 30 minutes pulling the logs and doing the math. Most contractors find that their actual missed-call rate is double what they thought, and once they see the dollar figure attached, the case for change makes itself. The data is already in your phone provider's dashboard. You just have to look at the right column.