Multiply your monthly call volume by your miss rate to get missed calls. Multiply those by your close rate to get lost bookings, then by average job value to get lost revenue. At 500 calls a month, a 22% miss rate, a 35% close rate and a $580 job, that is 38.5 bookings and $22,330 a month.
This is a four-step arithmetic model of the numbers you entered. It calculates the revenue attached to calls that were never answered, on the assumption that an unanswered call converts at zero. It is not a measurement of your business and it is not a forecast.
The calculation is deliberately boring, because a model you cannot follow in your head is a model you cannot argue with. Four numbers go in and one number comes out, and every step between them is visible on this page. Start with inbound calls per month. Multiply by your miss rate to get missed calls. Multiply those missed calls by the close rate you achieve on calls you do answer, which gives you the bookings you would have won if the phone had been picked up. Multiply that by the average value of one job, and you have the monthly revenue attached to calls nobody answered. The fifth input, recapture rate, is the honest part. It is not part of the loss calculation. It is a discount you apply on the way out, because no answering system, human or otherwise, converts every missed call. It ships at 60% because that is a conservative middle, and you should move it. Setting it to zero shows you raw exposure with nothing assumed about recovery at all.
TrainYourAgent quotes a figure of roughly $22,000 a month in missed-call exposure for a busy home-services business. That figure is this model, at these inputs: 500 inbound calls a month, a 22% miss rate, a 35% close rate on answered calls, and a $580 average job value. Run it yourself. 500 x 0.22 = 110 missed calls. 110 x 0.35 = 38.5 lost bookings. 38.5 x $580 = $22,330 a month, or $267,960 a year. That is a worked example of the formula, not a client result and not an average of anything. It is published here so the figure stops being an assertion and becomes something you can reproduce, disagree with, or replace with your own numbers in about thirty seconds. If your miss rate is 8% rather than 22%, the same model returns $8,120 a month, and that is the correct answer for you.
Estimating these two inputs is where most of the error lives, and both are usually already recorded somewhere you have access to. Miss rate: your VoIP or phone provider reports answered versus unanswered calls per period. Ask for the last full month, and ask for it split by hour of day if the system supports it. Most businesses discover their misses cluster into three windows: lunch, after close, and the fifteen minutes after a marketing send. Close rate: take the last ninety days of booked jobs that originated from a phone call, divide by the number of phone enquiries in the same period, and use that. If your CRM does not tag lead source, use a quarter you can reconstruct by hand rather than guessing across a year. Average job value: use the median invoice, not the mean, unless your job values are tightly clustered. A single large project will drag a mean far enough to make the whole model unusable.
The presets for HVAC, dental, legal, roofing and med spa are editable starting points, chosen to be plausible for the shape of each business. They are not survey data and not industry averages, and the tool labels them that way on screen. They differ because the businesses differ in structure. A roofing contractor has the highest miss rate on the list because the people who could answer are physically on a roof. A law firm has the lowest close rate because most callers are outside the practice area, but the highest single-matter value, so one miss costs more than a month of small tickets. A dental practice closes at a high rate because callers are usually booking a known appointment rather than shopping three quotes. The point of the presets is to save you typing and to show you how much the answer moves when structure changes. Overwrite every field with your own numbers before you take the output anywhere.
Compare it to the cost of fixing it, and be careful to compare monthly to monthly. A model that says $22,330 a month of exposure is only interesting next to what an always-on answer costs to run, which is a different calculation with its own published method. The useful next step is usually smaller than buying anything. Pull the hour-by-hour miss report first. If 70% of your misses are after hours, an after-hours-only answering path solves most of the problem at a fraction of the cost of a full-time solution. If your misses are spread evenly through the working day, you have a staffing or routing problem that no software fixes on its own. If the number is large enough to act on, the two tools linked at the bottom of this page price the fix from opposite directions: what an AI receptionist genuinely costs per minute, and what a faster first response is worth at your volume.
Not always. Some callers ring back, some leave a voicemail you return the same day. The model treats an unanswered call as worth zero because that is the clean upper bound, and then hands you a recapture-rate input to discount it. Set recapture to a low number if your callers are persistent.
There is no figure we can publish honestly here, because miss rates vary enormously by trade, staffing model and hours. Pull your own from your phone system. It is a one-report request and it will be more accurate than any benchmark anyone quotes you.
Median, in almost every case. A mean is dragged upward by a small number of large jobs, and since the model multiplies job value by lost bookings, that distortion is carried straight into the headline number.
No. The output is revenue, not margin. If you want a margin view, multiply the recoverable figure by your gross margin percentage before you compare it to the cost of any solution.