Key takeaways
- Answering services still exist, in fewer and larger firms: 1,035 US establishments in 2023, down from 1,487 in 2012[2][3].
- They pay off when a person has to hear the call now: emergencies, on-call patches and callers who will not leave voicemail.
- For bookings and price questions, a message-only service mostly moves voicemail somewhere else, because the caller still waits for a callback.
- Value the difference by running the after-hours calculator twice: once with on-the-spot booking, once with next-morning callbacks.
- Try the cheap fixes first: staggered lunches, a better greeting, fixed callback windows and forwarding only when nobody answers.
Still in business, in fewer and larger firms
Answering services still exist, and plenty of plumbers, clinics and law offices still send their nights and weekends to one. What has changed is the shape of the industry. The Census Bureau counted 1,487 telephone answering service establishments in 2012 with 43,820 employees[3]; by 2023 there were 1,035 establishments with 41,582 employees[2]. That is about 30% fewer locations with roughly the same headcount, so the typical service grew from about 29 people to about 40. Smaller local services have been closing or selling to larger ones.
The people who answer are thinning out too. The Bureau of Labor Statistics projects jobs for switchboard operators, including answering service operators, to fall from 35,400 in 2025 to 26,200 in 2035, a 26% drop[1]. The calls have not gone away. More of the routing, message delivery and simple questions are handled by software, and the human agents are kept for the calls that need judgment.
Most of what remains is still small: 758 of the 1,035 establishments had fewer than 20 employees in 2023[2]. For a buyer, that means two quite different products sold under one name. One is a small local service whose agents may come to recognize your regular callers. The other is a large operator whose agents answer for hundreds of accounts from a shared screen. Neither is wrong; they suit different calls.
What decides it: what the caller needs in the first minute
An answering service is worth it when the value of a call lies in a person hearing it now: deciding it is urgent, reaching your on-call person, or calming a caller who will not talk to a machine. It is worth much less when the caller needs something the agent cannot do, such as a booking, a quote or a specific answer about your work. Then the service turns a voicemail into a typed message, and the caller still waits for you.
| Call type | What the caller needs | What a message-only service delivers | Worth it? |
|---|---|---|---|
| Burst pipe, no heat or a lockout at night | Someone to reach the on-call tech now | A patch to the on-call phone, if patching is included | Yes, if the patch works |
| New customer wanting an appointment | A confirmed time | A message and a callback the next day | Weak |
| Price or availability shopper | A number or a yes | A message | Weak; often gone by morning |
| Existing customer with an update | To be heard and logged | A clear message with details | Fine |
| Patient calling a practice after hours | The on-call clinician, or reassurance | A patch or urgent page, under a signed agreement | Yes, with the agreement in place |
| Sales calls, robocalls, wrong numbers | Nothing | Billed minutes | No |
Sort last month's after-hours calls into these rows before asking for a quote. If most of the volume sits in the first and fifth rows, a service can pay for itself in emergencies alone; plumbers, for instance, often work nights and weekends and are frequently on call, according to BLS[4]. If most of it sits in the second and third rows, you would be paying to move voicemail somewhere else.
When it only moves voicemail somewhere else
There is a simple test for this. Measure the time from a caller hanging up to that caller being reached by someone who can help. With voicemail, the gap is however long your callback takes. With a message-only service it is the same gap plus a few minutes for the message to arrive. The service's real advantages are a cleaner message and a caller who heard a person. Both setups then hit the same wall: in Pew's survey, 67% of adults said they let calls from unknown numbers go to voicemail and check it later, and only 19% said they generally answer[5].
after-hours calls answered by the service
from new customers, each turned into a message
reached on the first callback the next morning
still needed the job and booked
Example figures for illustration; your own message log will show where your callers drop out.
Example: a dental practice whose evening callers mostly want to book a cleaning gains little from a message service. The patient could have left the same request on voicemail, and the front desk calls back at 8 a.m. either way. The same practice gains a great deal if its on-call dentist is reached when a patient calls at 9 p.m. with a swollen jaw. That needs a signed business associate agreement first, because a practice may only let an outside service handle patient information on its behalf after documenting written assurances that it will be safeguarded[6].
Callbacks also cost staff time that rarely shows up in the comparison. Example: twenty minutes every morning spent dialing back overnight messages and leaving voicemails is about seven hours a month. At the $36.73 an hour employers paid on average in total compensation for office and administrative staff in June 2026[7], that is roughly $265 a month of someone's time before a single job is booked.
Putting a number on it
The calculator below uses a fictional four-truck plumbing company that gets 14 after-hours calls a week, half of them from new customers. The average job value is the American Housing Survey's median homeowner spend on plumbing fixture projects[8].
| Input | Example |
|---|---|
| After-hours calls per week | 14 |
| From new customers (%) | 50% |
| Share of new callers who book | 45% |
| Average job value | $500 |
| After-hours calls a month | 61 |
| From new customers | 30 |
| Bookings if answered | 14 |
| Monthly after-hours value | $6,820 |
How it is worked out: After-hours calls a week × 4.33 weeks × new-customer share × booking rate × average value.
Assumptions
- Average job of $500: the median homeowner spend on plumbing fixture projects in the 2023 American Housing Survey.[8]
- A 45% booking share assumes new callers are booked or reached on the call; most people screen unknown numbers, so next-morning callbacks book fewer.[5]
- The 14 weekly calls and 50% new-customer share are invented for a four-truck company; use your own after-hours log.
Example only: about $6,820 a month of work behind after-hours new-customer calls when they are booked on the spot. Lower the booking share to see what next-morning callbacks leave.
Now run it again with the booking share cut from 45% to 25%, a stand-in for what happens when a new caller gets a message taken instead of a booked slot and has to be reached again in the morning. In this example the value falls from about $6,820 to about $3,790 a month. Both percentages are assumptions; replace them with your own. The gap between the two runs is the cost of relaying instead of resolving, and it is the number to hold against any quote. The after-hours plumbing example works the same model for a five-truck company, and the plumbing guide to an AI phone agent versus a live answering service compares the two on emergency triage.
Run the relay test on your own line
You do not need a vendor's trial to find out which side of the line you are on. Two weeks of your own records will do it, one with your current setup and one with whatever you are considering. Keep the log on paper or in a shared sheet, and fill it in the same morning, while the details are fresh.
Log every out-of-hours call
Time of the call, the caller's number, and whether a message or voicemail was left. Your phone system's call history fills most of this in.
Note when the message reached you
For voicemail that is the time someone first listened to it; for a service, the time the email, text or portal message arrived.
Note when the caller was actually reached
Not when someone dialed back, but when a person on your side spoke to the caller. Count voicemail tag as not yet reached.
Record the outcome
Booked, sent elsewhere, not needed, or lost. Ask lost callers who agree to talk what they did instead; most will tell you.
Compare the two weeks
Median time from hang-up to first real conversation, and the share of new callers who booked. Those two numbers settle most debates.
Read the result plainly. If the service shortens the time to a real conversation only by the minutes a message takes to arrive, it is relaying, and the booking share will barely move. If it shortens it to zero for the calls that matter, because the agent patched an emergency through or booked the appointment there and then, the booking share moves, and the service is earning its fee. Keep the log either way. It is the evidence you will want when the contract comes up for renewal.
Cheaper fixes to try first
Before signing anything, check whether a smaller change closes most of the gap. Each of these costs little and can be tried in a week.
- Fix lunch. If last month's missed calls bunch between noon and 1, stagger breaks so the phone is always covered by someone.
- Rewrite the greeting so callers know when they will hear back and what to leave; the voicemail greeting examples include scripts for after hours and busy lines.
- Return messages in fixed windows, such as 8:30 and 1:00, with a callback script that confirms the details in one call.
- Forward on no-answer only. Overflow cover protects the busiest hours without paying for quiet ones.
- Hand routine calls to software. An SI Employee can book routine requests into a connected calendar and transfer urgent callers to the numbers you approve, leaving people for the calls that need a person.
If none of these move the number, a service is worth pricing. The guide to answering services for small business covers what the base plan includes and the contract terms that set the bill. If your calls run into the hundreds a day, read call center vs answering service first, because at that volume a dedicated team starts to make sense.
Where the answer usually lands
- Worth it: on-call trades, property managers and medical offices with real night emergencies, provided patching to a person is included and tested.
- Worth it: owners who cannot pick up during the day, in court, on a roof or with a patient, whose callers will not leave voicemail.
- Not worth it: lines where most callers want bookings or prices, unless the service can book against your real availability.
- Not worth it: low volume where most callers are existing customers who are happy to leave a message.
Whichever way you land, write down why. An owner who signs because a competitor uses a service, or cancels because one message went astray, usually ends up switching again within a year. An owner who signs because emergencies were going unanswered, and checks that they no longer are, knows when the arrangement has stopped paying.
The honest answer for many small businesses is a mix: a person or service for the emergencies, something that books for the routine, and voicemail for the rest. For a side-by-side with software that finishes the task on the call, see SI Employee vs answering service. For live agents against AI answering in general, read AI vs live answering service.
Frequently asked questions
Are answering services still a thing?
Yes. The Census Bureau counted 1,035 answering service establishments in 2023, though the number has been falling as firms consolidate and software takes over routine calls[2][3].
What do callers get from an answering service that voicemail does not give them?
A person who can take detail, reassure them and, if it is set up, reach your on-call staff. What most plans do not give them is a booking or a firm answer to a specific question.
How do I know if my calls are urgent enough to justify a service?
Sort a month of after-hours calls into emergencies, bookings, questions and noise. A handful of emergencies a month can justify coverage if each one is a large job or a safety issue; if most calls are bookings, look for something that can book.
Is an answering service cheaper than hiring someone?
For after-hours and overflow cover, usually, because you pay for minutes rather than a shift. For daytime cover, compare the quote with the hours a part-time hire would actually work, not with a full-time salary.
Can an answering service hurt my business?
It can, if agents give wrong information, mishear phone numbers or sound like they are reading a script for a hundred companies. Test calls before launch and a weekly read of messages catch most of it.
References
U.S. Bureau of Labor Statistics (2025). Occupational separations and openings, projected 2025-35 (opens in a new tab) ↑
U.S. Census Bureau (2023). County Business Patterns 2023, U.S. detail file (NAICS 561421, telephone answering services) (opens in a new tab) ↑
U.S. Census Bureau (2012). County Business Patterns 2012, U.S. detail file (NAICS 561421, telephone answering services) (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2026). Plumbers, Pipefitters, and Steamfitters: Occupational Outlook Handbook (opens in a new tab) ↑
Pew Research Center (2020). Most Americans don't answer cellphone calls from unknown numbers (opens in a new tab) ↑
Legal Information Institute, Cornell Law School (2026). 45 CFR 164.502: Uses and disclosures of protected health information, general rules (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2026). Employer Costs for Employee Compensation, Table 4: private industry workers by occupational and industry group, June 2026 (opens in a new tab) ↑
U.S. Census Bureau and U.S. Department of Housing and Urban Development (2023). Home Improvements (2023 American Housing Survey infographic) (opens in a new tab) ↑