Key takeaways
- Menus help multi-trade and multi-location companies route calls; they usually hurt single-trade shops with one office.
- If you use one, keep it to two options plus an emergency path, and put the most common choice first.
- Most callers are on mobile phones: 98% of US adults own a cellphone and 91% a smartphone[1], so menus must be short and easy to follow hands-free.
- Test before and after: measure hang-ups during the menu, misrouted calls and time to a person.
The case for and against menus
An automated phone menu, often called an IVR or auto attendant, can route calls to the right team before anyone answers. It can also add 20 seconds of friction to every call and lose impatient callers before they ever hear a person. Whether a menu helps depends almost entirely on how many different places a call could need to go.
For a one-location HVAC shop with two CSRs, every call goes to the same people, so a menu adds delay without adding value. For a company with plumbing, HVAC and electrical divisions in three cities, a short menu can save every caller a transfer. Most companies sit somewhere in between, which is why the decision deserves a test rather than a guess.
Decision table
| Your situation | Menu? | Why |
|---|---|---|
| One trade, one office, under three people answering | No | Everyone answers the same calls; the menu only adds delay |
| One trade, after-hours on-call routing | Minimal | A single prompt for emergencies after hours can help |
| Several trades with separate teams | Yes, two or three options | Callers reach the right booking team without a transfer |
| Several locations with separate schedules | Yes, or separate numbers | Route by location, or give each location its own number |
| High share of existing-job status calls | Maybe | An option for status calls can free CSRs for new bookings |
| Many Spanish-speaking callers | Yes, a language option | Routes callers to a bilingual CSR from the start |
When menus hurt
- Urgent callers who just want a person. Someone with water coming through the ceiling will not wait through four options.
- Long lists. Callers forget the first option by the time they hear the fourth, and press zero or hang up.
- Menus that reset. Pressing the wrong number and being sent back to the start is the fastest way to lose a call.
- Menus that do not match the business. An option for “sales” confuses a homeowner who wants a repair; plain words such as “book a repair” work better.
- Speech recognition that fails. Voice menus struggle with accents, background noise and callers on speakerphone in a car.
Mobile use makes length matter. Nearly all US adults own a cellphone, 91% own a smartphone and 16% are smartphone-only internet users without home broadband[1]. Many service calls come from cars, driveways and job sites, where a long spoken menu is hard to follow and impossible to write down.
A two-option design that works
“Thanks for calling [Company]. If you smell gas or there is a fire or medical emergency, hang up and call 911. To book a repair or estimate, press 1. If we're already scheduled to visit you, press 2. Or stay on the line and we'll answer.”
- Safety first: one sentence, before any option.
- The most common choice first, as option 1.
- No more than two numbered options for most companies; three at most for multi-trade firms.
- Staying on the line always reaches a person, so callers who do nothing are never lost.
- No marketing in the menu and no request to visit the website.
Multi-trade companies can replace option 1 with trades, such as “For heating and cooling, press 1. For plumbing, press 2. For electrical, press 3,” and move existing-job calls to option 4 or to the stay-on-the-line path. Multi-location companies are often better served by a separate number per location than by a location menu.
After hours: one prompt, not a maze
After hours, the menu has one job: separate emergencies from everything else. A single prompt such as “If this is an emergency, press 1 to reach our on-call technician; to book a regular visit, press 2” is enough. Define emergencies in words callers understand: no heat, no cooling in extreme heat, active leaks, a door stuck open or burning smells. Make sure option 2 leads somewhere that can book, whether an answering service, an online booking link or an AI employee, rather than a voicemail box nobody checks until morning.
Alternatives to a menu
Many of the problems companies try to solve with a menu have simpler fixes that do not add a step for every caller.
- Separate numbers: give each location, trade or the billing team its own number, printed where those callers will see it, such as on invoices for billing.
- Smart routing without prompts: route known customers to their usual team using caller ID, and new callers to the booking team.
- Ring groups: let two or three people's phones ring at once, so the first free person answers, instead of asking the caller to choose.
- Time-based routing: send calls to the on-call phone or overflow service automatically after hours, with no prompt at all.
- A booking link in your missed-call follow-up and on your website for routine work, so fewer routine calls reach the queue.
Try these first if you are thinking about a menu to fix a specific problem, such as billing calls clogging the booking line. Often one change removes the need for a menu entirely.
Writing the menu wording
Menu wording is read once, out loud, to someone in a hurry. Write it for the ear. Use the words customers use, such as “repair” and “appointment,” rather than internal labels like “service department” or “CSR team.” Put the action before the number: “To book a repair, press 1,” not “Press 1 for repair booking.” Keep each option under ten words. Record it in a clear, warm voice at a normal pace, and test it by calling from a mobile phone on speaker in a moving car. If the wording works there, it will work anywhere.
Self-service options inside the menu
BLS notes that self-service systems, social media and mobile apps let customers do simple tasks without a representative[2]. A menu can offer some of this: confirming an appointment, hearing the arrival window or getting a text link to reschedule. Use self-service only for tasks that genuinely need no judgment. Anything involving a new problem or a safety question should reach a person or a trained answering system quickly.
If you use an AI system or speech recognition in the menu, tell callers plainly what they are talking to. Half of US adults say they are more concerned than excited about the increased use of AI in daily life[3], and surprises erode trust faster than a clear introduction does.
How to test menu abandonment
Measure a baseline week
Record total calls, calls that hung up during the menu, calls that reached a person, median time to a person and the share of calls transferred after answering.
Change one thing
Remove options, reorder them or turn the menu off during open hours. Change only one element per test.
Measure the test week
Use the same filters and compare. Avoid comparing a normal week with a heat wave.
Check misroutes
Ask CSRs to tag calls that arrived at the wrong team. A menu that cuts transfers but loses callers at the prompt is not a win.
Decide and document
Keep the version with the higher share of callers reaching the right person quickly. Record the result so the next manager does not undo it.
Example: removing three options
Example: a fictional company with HVAC and plumbing divisions had a five-option menu: sales, service, billing, existing appointments and “all other questions.” Its call log showed many callers hanging up during the menu and frequent transfers between service and sales. It replaced the menu with “heating and cooling, press 1; plumbing, press 2; or stay on the line,” moved billing to a separate number printed on invoices and added a safety sentence. Hang-ups during the menu fell sharply in the first month and CSRs reported far fewer transfers.
Who handles the routed calls
A menu is only as good as the people it routes to. In multi-trade shops, dispatchers often answer as well as schedule; building equipment contractors employed 14,490 dispatchers in May 2025[4]. If a menu option sends calls to a dispatcher who is also on the radio with techs, that option will ring out at the worst moments. Give every option a backup answer point.
Track menu results alongside your queue numbers. For how long callers wait once routed, see how long callers will wait on hold, and for what happens on the call itself, how to review recorded calls and coach CSRs.
Frequently asked questions
Does a small home service company need a phone menu?
Usually not during open hours. If everyone who answers handles the same calls, a menu only adds delay. A single after-hours prompt to separate emergencies can still help.
How many options should a phone menu have?
Two for most companies, three at most for multi-trade firms, plus a safety sentence and a stay-on-the-line path that always reaches a person.
Should emergencies be the first option?
Put the safety instruction first, telling callers with gas smells, fires or medical emergencies to call 911. For urgent service such as no heat or leaks, a clear emergency option after hours works well.
How do I know if my menu is losing calls?
Check your phone system for calls that hung up during the menu, compare a week with and without the menu, and watch misroutes and time to a person.
References
Pew Research Center (2026). Americans' Use of Mobile Devices in 2026 (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). Customer Service Representatives: Occupational Outlook Handbook (opens in a new tab) ↑
Pew Research Center (2025). How Americans View AI and Its Impact on People and Society (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). Occupational Employment and Wage Statistics, May 2025, series OEUN000000023820043503201 (Dispatchers, except police, fire, and ambulance, in Building Equipment Contractors) (opens in a new tab) ↑