TL;DR & Quick Summary
Most businesses know they miss calls. Very few know what that costs, which is why the problem stays unfixed — it never gets a number attached, so it never competes for budget against things that do.
This is the method for producing that number, using data you already have.
The formula:
Missed calls per month × prospect share × booking rate × average job value = monthly lost revenue
Where to get each input — your phone system and your accounts, not a guess
Benchmark ranges by industry so you can sanity-check your result
The two inputs everyone gets wrong, both of which inflate the answer
Key Takeaway: The number is usually larger than owners expect and smaller than vendors claim. Calculating it yourself is the only way to know which side you are on.
Get Started: Want this run against your real call data before you commit to anything? Schedule a Strategy Call with Cogniq AI or explore our no-show and missed call prevention services.
Why This Number Stays Invisible
Missed calls are uniquely easy to ignore, for three reasons.
There is no record of the loss. A missed sale in a CRM leaves a trace. A missed call leaves a voicemail nobody listens to, or nothing at all. The revenue that did not arrive is not recorded anywhere, so it never appears in a report.
Recall is biased toward success. Owners remember the calls they answered, because those became conversations and jobs. The unanswered ones are, by definition, not memorable.
The cost is distributed. Two missed calls on a Tuesday feels like nothing. Forty a month at a meaningful job value is a hire's worth of revenue.
Attaching a figure changes the conversation from "we should probably answer more calls" to a decision with a comparable cost on the other side.
The Four Inputs
Input 1: Missed calls per month
Where to get it: your phone provider's call log. Every mainstream business phone system reports answered and unanswered volume.
Count as missed anything that rang out, went to voicemail, was abandoned in a queue, or hit a busy signal. Pull three months and average — single months mislead badly in seasonal trades.
Split it two ways while you are there:
- Business hours vs after hours. Different problems, different fixes.
- By day and time. Missed calls cluster. Most businesses find a small number of predictable windows carrying most of the loss.
Input 2: Prospect share
Where to get it: a sample of 50–100 calls, checked by hand against your customer list.
Not every inbound call is a potential job. Real inbound volume includes existing customers, suppliers, spam, wrong numbers, and recruiters. This is the input that most distorts the calculation, because treating every missed call as a lost lead can inflate the result several times over.
Typical ranges once cleaned:
| Business Type | Prospect Share of Inbound |
|---|---|
| Trades and home services | 45–65% |
| Dental and medical practices | 25–40% |
| Professional services (legal, accounting) | 30–50% |
| Salons, spas, clinics | 40–60% |
Practices skew low because a large share of inbound is existing patients rescheduling. Trades skew high because most calls are new work.
Input 3: Booking rate
Where to get it: your CRM or job management system, counting answered prospect calls only.
This is the second input people get wrong. Use the rate at which answered enquiries become booked work. Do not use your overall close rate, which includes quotes and follow-ups that started elsewhere.
A common range is 30–50% for service businesses with immediate need, lower where the purchase involves comparison or a quote.
Input 4: Average job value
Where to get it: your accounts. Total revenue divided by jobs completed, over the same period.
Two adjustments worth considering:
- Use first-job value, not lifetime value, unless you can substantiate retention. Lifetime value is where these calculations become fantasy.
- Use the median, not the mean, if you have a handful of large jobs. One outlier contract can distort the average beyond usefulness.
Worked Example
A three-van plumbing business:
| Input | Value | Source |
|---|---|---|
| Missed calls per month | 60 | Phone system, 3-month average |
| Prospect share | 55% | Sample of 80 calls |
| Booking rate | 40% | Job management system |
| Average job value | $800 | Accounts, median |
60 × 0.55 × 0.40 × $800 = $10,560 per month
Roughly $127,000 a year in work that rang and was not answered.
Now split it:
| Segment | Missed Calls | Monthly Value |
|---|---|---|
| Business hours (on a job, on another call) | 38 | ~$6,690 |
| After hours and weekends | 22 | ~$3,870 |
That split matters more than the total, because the two halves have different solutions. The after-hours figure is the cleanest — those calls currently reach voicemail, so anything that answers them is pure recovery. The business-hours figure competes with work already being done.
If Your Phone System Does Not Report This
Some VoIP providers and most traditional lines expose less than you need. Three fallbacks, in order of preference:
Ask your provider directly. Call data exists even when the dashboard does not surface it. Support can usually export a call detail record covering answered, unanswered and abandoned calls for a date range.
Use your mobile carrier's log, if calls forward to a phone. It will show missed calls, though not queue abandons.
Run a two-week manual count. Crude, but better than guessing. Log every call that rang out, double it for a monthly estimate, and make sure the fortnight was not a holiday period.
The Blind Spot Worth Checking
Most businesses count sequential missed calls and never look at simultaneous ones. If two people call at 9:15 and one person answers one line, the second caller hears a busy tone or rolls to voicemail — and on many systems that second call is never logged as missed at all, because it was never presented to anyone.
This matters because concurrent calls cluster exactly when demand is highest. If your call log shows a suspiciously flat distribution across the day while your team insists mornings are chaotic, simultaneous calls are the likeliest explanation, and your real missed number is higher than the report says.
Ask your provider specifically whether concurrent-call rejections appear in the report. If they do not, add a week of manual observation to catch them.
Benchmarks to Sanity-Check Your Result
Use these to check whether your inputs are plausible — not as substitutes for measuring.
| Industry | Typical Unanswered Rate | Typical First-Job Value |
|---|---|---|
| Plumbing, HVAC, electrical | 25–40% | $400–$1,200 |
| Dental practice | 15–30% | $200–$600 first visit |
| Legal (small firm) | 20–35% | $1,500–$5,000 matter |
| Salon and spa | 20–35% | $60–$200 |
| Veterinary | 15–30% | $150–$400 |
| Auto repair | 25–40% | $300–$900 |
Ranges are indicative and vary widely by market and operating model. If your result sits far outside them, check your prospect share first — it is almost always the culprit.
The Two Mistakes That Ruin the Calculation
1. Counting every missed call as a lost lead. The single most common error, and it can inflate the number by two or three times. Vendor calculators frequently do this by default because it produces a more persuasive figure. Sample your calls and use a real prospect share.
2. Using lifetime value instead of first-job value. Multiplying by a lifetime figure you cannot substantiate turns a business case into a wish. If retention is genuinely strong and you can evidence it, model it separately as an upside case — do not bake it into the headline number.
A third, subtler one: assuming a 100% answer rate is achievable. It is not, and it is not worth pursuing. Model recovering the gap to roughly 90–95%, not to perfection.
What to Do With the Number
Once you have it, you can compare it against real options rather than argue about it.
| Option | Typical Monthly Cost | Covers |
|---|---|---|
| AI voice agent | $150–$400 | All hours, unlimited concurrent calls |
| Human answering service | $1,700–$2,100 | All hours, priced per call or minute |
| Additional staff member | Salary plus overhead | Business hours only, one call at a time |
| Do nothing | $0 | Voicemail |
The comparison is not AI against a person. It is whichever option you choose against voicemail, because voicemail is what currently answers.
If your calculated loss is a few hundred dollars a month, the honest answer may be that better call routing or a shared answering rota solves it and no technology is warranted. If it runs into five figures, the decision makes itself. Either way you now have grounds for it.
For what the options actually cost to run, see our AI voice agent cost teardown and our AI receptionist pricing comparison. For the wider exercise of finding which processes are worth automating at all, our AI automation audit playbook covers the method.
Measure Again in 90 Days
Whatever you change, re-run the same four inputs a quarter later. Two things to watch:
- Answer rate should move first and most visibly.
- Booking rate on newly answered calls is the honest test. Calls answered by a system that cannot book are not worth the same as calls answered by one that can, and this is where a badly integrated deployment shows up in the numbers.
If answer rate improved but booked jobs did not, the problem was never the phone — it was what happened after it was picked up.
Conclusion
The reason missed calls stay unfixed is that they are invisible in every report a business actually looks at. The fix is not a tool. It is a number: four inputs, all of which you already have, multiplied together.
Do that arithmetic before you take a single vendor call. You will either discover a problem worth solving, or discover that yours is smaller than the solutions being sold to you — and both outcomes are worth the twenty minutes.
Schedule a Strategy Call with Cogniq AI and we will run these numbers against your actual call log, including when the honest answer is that you do not need what we build.
