Key takeaways
- Conversion rate = bookings ÷ bookable calls. Missed calls stay in the denominator; vendors, spam and service calls come out.
- Split it into reach rate and booking rate to see whether the leak is speed, script or availability.
- Callbacks recover few missed callers: 67% of US adults let unknown numbers ring and rely on voicemail[2].
- Availability is a conversion problem too: new dental patients waited 16.0 days on average for a first visit in 2025[3].
- There is no reliable cross-industry benchmark. Measure four weeks, change one thing, measure four more.
Same calls, different results
Two businesses can take the same 200 calls in a month and book very different numbers of jobs. The difference rarely shows in the call log, which records only answered or missed. It shows when you follow each call to its end: booked, quoted, promised a callback, told the earliest slot was three weeks away, or never reached anyone at all.
Your inbound call conversion rate is the share of bookable calls that end in a booking. Measured properly, it tells you which of three problems you have. Speed: callers do not reach anyone. Script: they reach someone who never asks for the booking. Availability: there is nothing to offer them when they do. Each has a different fix, and the wrong fix spends money without moving the number.
First conversations decide more than owners think. In Zillow's 2026 survey, 44% of recent home buyers hired their agent after contacting them only once, and another 25% after two contacts[1]. Real estate is one market, but many service buyers behave the same way: the first business that holds a useful conversation gets the work.
Choose the denominator before you count anything
Most conversion numbers go wrong at the bottom of the fraction. Divide bookings by all calls and vendors, wrong numbers and billing questions drag the rate down. Divide by answered calls only and the missed calls vanish from the problem. The fix is to give every call one outcome code and decide in advance which codes count.
| Outcome code | In the denominator? | What it tells you |
|---|---|---|
| Booked | Yes | A job, appointment or consultation is on the calendar |
| Request logged | Yes | Details taken; staff must confirm a time |
| Quoted, not booked | Yes | Price or scope stopped the booking; review the call |
| Callback promised | Yes | No booking yet; time how long the callback takes |
| Missed or voicemail | Yes | The caller never reached anyone: a speed problem |
| No suitable slot | Yes | An availability problem, not a sales problem |
| Out of area or not offered | No, track separately | A targeting problem in your ads or listings |
| Existing job, billing, status | No | Service calls; measure them, but on their own |
| Vendor, spam, wrong number | No | Exclude |
With those codes the formula is short: bookings divided by bookable calls, where bookable means every call in the first six rows. Count a logged request as a booking only once staff confirm it, or you will flatter the rate. Your own call disposition codes can be shorter or longer than this list, but they must separate a missed call from a lost one.
Keep new and existing callers apart as well. Existing customers usually book more readily, and a strong repeat-customer number can hide a weak new-customer one for months.
Decide who codes each call and when. The person who took the call should code it before picking up the next one, because by the end of the day a callback promise and a quote blur together. For calls handled by an answering service or software, code from the call summary the same day. A weekly 20-minute pass to fill the gaps keeps the data clean enough to trust.
Measure it as a funnel, not a single percentage
A single rate hides where callers drop. Break it into stages and multiply them back together. The example below is a fictional six-person home services office over one month.
inbound calls
bookable calls after removing vendors, spam and service calls
reached a person or the answering system
booked, or confirmed from a logged request
Example figures: reach rate 79%, booking rate on reached calls 61%, overall conversion 48%.
Reach rate times booking rate gives overall conversion: 79% × 61% is about 48% in the example. Now the two levers are visible. Lifting reach to 95% at the same booking rate takes overall conversion to 58%, about 13 more bookings a month. Lifting the booking rate to 70% at today's reach takes it to 55%, about 9 more. Which to work on first depends on which is cheaper to move, and the log usually makes that obvious.
If you buy search ads, do not let the ad platform's conversion count stand in for this. Google Ads, for instance, counts a call to a website number as a conversion when it lasts longer than a minimum length the advertiser sets[4]. A long call can end with no booking, and a short one can book a repeat customer. Use the ad count to compare campaigns and your own outcome codes to measure the phones.
Find where callers drop
Read a month of outcome codes and the pattern usually points to one of three leaks.
| What the log shows | Likely leak | First fix |
|---|---|---|
| Bookable calls missed or sent to voicemail, clustered at lunch or after 5 pm | Speed | Cover those hours: stagger breaks, add overflow or after-hours answering |
| Calls reached but ending in a callback promise | Script | Ask for the booking on the first call and offer two specific times |
| Quotes given, few booked, calls under two minutes | Script | Explain the visit or diagnostic fee and what it covers before the price |
| No suitable slot coded several times a week | Availability | Hold some same-week capacity for new customers |
| Many out-of-area callers | Targeting | Tighten ad areas and state your service area in listings |
| Callbacks taking more than an hour | Speed | Return calls within minutes and leave a voicemail with a time |
Speed is the most common leak and the easiest to see. A missed call is rarely recovered by ringing back later: in Pew's survey, 67% of US adults said they usually let calls from unknown numbers ring and check voicemail if one is left, and 14% ignore voicemail as well[2]. Your callback comes from a number they have not saved. If you must call back, leave a short voicemail with the business name, the reason and a specific time you will try again. Google also says missed calls may count against responsiveness in Local Services Ads ranking[5], so a speed leak can shrink next month's call volume as well as this month's bookings.
Example callback voicemail: 'Hi, this is Dana at Northside Heating, returning your call about the furnace. I'll try you again at 2:15. If that's a bad time, call this number back before 6 and ask for me, and we'll get a visit booked.'
Availability is the leak owners fix last, because it feels like a capacity question rather than a phone question. It is both. The American Dental Association's practice survey shows how quickly it can change: the average wait for a new patient's first appointment at a general dentist doubled from 7.9 days in 2019 to 16.0 days in 2025, while existing patients waited 9.3 days[3].
Source: American Dental Association Health Policy Institute, 2025[3]
Show the data
| Item | Value |
|---|---|
| New patient, 2019 | 7.9 days |
| New patient, 2025 | 16 days |
| Patient of record, 2025 | 9.3 days |
A new caller offered a slot two weeks out has plenty of time to ring someone else. When no suitable slot shows up in your codes, holding a few same-week openings for new customers can lift conversion more than any script change. Buyers are already struggling to find help: 53% of homeowners surveyed for Angi's 2025 home spending report said they had difficulty finding pros[6]. The business that can say Thursday morning gets the job.
Fix the leaks in this order
Close the speed gap
Cover the hours your log shows as missed. Forward overflow and after-hours calls to someone or something that answers, whether a person, an answering service or an SI Employee, and measure the reach rate again after two weeks.
Make the booking ask
Every bookable call ends with a time offered. Train the line: 'I can have someone there Thursday between 8 and 11 or Friday between 1 and 4. Which is better for you?'
Handle price questions on the call
Give the visit or diagnostic fee and what it covers, then offer the time. A clear fee and a slot give the caller something to say yes to; 'it depends' gives them a reason to keep calling around.
Protect same-week capacity
Keep a few slots each week for new customers until 48 hours before, then release them to existing customers.
Put a clock on callbacks
Every promised callback gets a deadline. Measure the minutes from the missed call to the return call, not just whether it happened.
Review ten calls a week
Listen to five that booked and five that did not. The difference is usually one sentence, and it is usually the same sentence.
Change one thing at a time. If you add evening coverage and a new script in the same week, you will not know which one moved the number, and you will keep paying for both.
What a bookable call should look like
- Caller
- System
- SI Employee
- Your team
01Caller
Caller rings
A new customer calls the business number about a problem or a service.
02System
Answered in a few rings
Staff pick up, or the call forwards on overflow or after hours.
03SI Employee
Need and details taken
Name, callback number, address or reason, and how urgent it is.
04SI Employee
Two times offered
Booked into a connected calendar, or logged as a request with preferred times.
05System
Summary to the office
Summary by email; recording and transcript kept in the account.
06Your team
Requests confirmed
Staff confirm logged requests the same day and code the outcome.
Example flow. Steps three and four can be handled by staff or by answering software; the outcome code is what makes the call measurable.
Whoever handles the middle steps, every bookable call should end with a booking or a dated next step. If you use software for overflow, check that it can take a request with preferred times when no calendar is connected, and that every call reaches you as a summary you can code. The walk-through of how an AI receptionist handles a call and the questions to ask before you buy one cover what to test.
Targets, and how long to wait for results
There is no trustworthy cross-industry benchmark for inbound call conversion. Published figures mix call types, count answered calls only, or come from companies selling the fix. The useful benchmark is your own baseline: measure four weeks, change one thing, measure four more. Reach rate responds within a week of new coverage; booking-rate changes from a new script need three or four weeks of calls before one bad day stops swinging the result.
Watch for changes that move the rate for reasons that have nothing to do with the phones. A price increase, a new ad campaign that pulls in out-of-area callers, or a seasonal shift in call mix will all move conversion without anyone answering differently. Note those dates on the tracking sheet so the phones are neither credited nor blamed for them.
It is worth the effort. In the Federal Reserve Banks' 2024 survey, 57% of small employer firms named reaching customers and growing sales as an operational challenge, the most common one[7]. Callers who already rang you are the cheapest customers you will ever reach, and conversion is the measure of how many of them you keep.
Then put a value on each point of conversion. An extra booking is worth its gross profit today and, for repeat-service businesses, far more over time; customer lifetime value shows how to put a number on that, and the software payback period guide shows how to judge whether a tool that lifts the reach rate earns its keep. For reach and booking worked through on a busy front desk, see the dental practice example. Industry guides go further for home services call funnels and new-patient calls.
Frequently asked questions
What is a good inbound call conversion rate?
It depends on call mix, price and capacity, so compare against your own baseline rather than a published average. Track new-customer calls separately; that is the figure that moves when you fix speed or script.
Do missed calls count in the conversion rate?
Yes. Missed bookable calls belong in the denominator. Leaving them out makes the phones look better than they are and hides the speed leak entirely.
How do I measure conversion without call tracking software?
Export the call log from your phone system, add an outcome column, and match caller numbers to bookings in your calendar or job system once a week. A spreadsheet is enough for a business taking a few hundred calls a month.
Should a call that books later by email count?
Count it if the booking follows within your matching window, often 7 to 14 days, and code it as booked after the call. The call did the work; the email confirmed it.
Is call length a good proxy for a converted call?
Only a rough one. Ad platforms use a minimum call length because they cannot see your calendar, but long calls can end without a booking and short ones can book a returning customer. Outcome codes are more reliable.
References
Zillow Research (2026). Buyers: Results from the Consumer Housing Trends Report 2026 (opens in a new tab) ↑
Pew Research Center (2020). Most Americans don't answer cellphone calls from unknown numbers (opens in a new tab) ↑
American Dental Association Health Policy Institute (2025). Income, Gross Billings, Expenses, Characteristics, Selected Staff Wages: Selected 2025 Results from the Survey of Dental Practice (Table 22) (opens in a new tab) ↑
Google (2026). Set up tracking calls to a phone number on a website (Google Ads Help) (opens in a new tab) ↑
Google (2026). About ad rankings (Local Services Ads Help) (opens in a new tab) ↑
Angi (2025). Angi's 2025 State of Home Spending Report Highlights (opens in a new tab) ↑
Federal Reserve Banks (Small Business Credit Survey) (2024). 2025 Report on Employer Firms: Findings from the 2024 Small Business Credit Survey (opens in a new tab) ↑