Key takeaways
- Answer rate is calls answered live by a person or service divided by real inbound calls; missed-call rate is the rest.
- Clean the data first: remove internal calls, spam and hang-ups of a few seconds, and count repeat callers once per day.
- Track time to first callback for missed calls, because sellers increasingly contact just one agent[1].
- Set targets by hour of day and call type, using four weeks of your own data as the baseline.
Why these two numbers
Most brokerages know their closed volume to the dollar and have no idea how many calls they miss. Yet the phone is still where many sellers and buyers make first contact, and first contact is decisive. In Zillow’s 2026 survey, the share of agent-using sellers who contacted only one agent rose to 67%[1], and three in four rated an agent’s initial responsiveness as very or extremely important[1]. If the one call a seller makes goes unanswered and unreturned, there is no second chance to lose.
Source: Zillow Research, 2026[1]
Show the data
| Item | Value |
|---|---|
| 2025 | 62% |
| 2026 | 67% |
Answer rate and missed-call rate turn that risk into something you can see each week. They are simple to calculate, and once you have them you can decide whether the problem is staffing, routing or follow-up.
Definitions you can apply consistently
Write the definitions down and keep them fixed. A metric that changes meaning every month cannot show progress.
| Metric | Formula | Notes |
|---|---|---|
| Inbound calls (clean) | All inbound calls minus internal, spam and calls shorter than a set threshold | Pick one threshold, such as five seconds, and keep it |
| Answer rate | Calls answered live ÷ clean inbound calls | Count answers by agents, staff, a service or an SI Employee as answered |
| Missed-call rate | Calls not answered live ÷ clean inbound calls | Includes voicemail, abandoned in queue and no answer |
| Voicemail rate | Missed calls that left a voicemail ÷ missed calls | Shows how many callers bother to leave a message |
| Callback rate | Missed calls returned ÷ missed calls | Measured within a set window, such as same business day |
| Time to first callback | Median time from missed call to first return attempt | Report by call type and by hour of the original call |
| Unreturned missed calls | Missed calls with no return attempt within the window | The number to drive toward zero |
Where to find the data
Almost every business phone system can export a call log. The fields you need are the same everywhere, even if the names differ.
- Call start time, in your local time zone.
- Direction (inbound or outbound) and the number dialed, so you can separate the main line, listing lines and agent lines.
- Caller number, to spot repeat callers and outbound callbacks to the same number.
- Disposition: answered, voicemail, abandoned, no answer, forwarded.
- Duration, to filter out very short calls and spam.
- Who answered, so you can see how much load each person or service carries.
Agents who use personal cells for client calls are the gap. Their calls do not appear in the office system. If most of your listing traffic goes to personal numbers, the office answer rate tells you little. One fix is to use business numbers on listings and signs that route through a system you can report on; another is to ask agents to export their own call history monthly. Fewer than half of REALTORS use a CRM[2], so the phone log may be the only record you have.
Clean before you calculate
Remove internal calls
Calls between extensions or from your own staff cells are not customer demand.
Filter spam and robocalls
Use your carrier’s spam labels and a minimum duration threshold. Keep a list of known spam numbers.
Handle quick hang-ups
Decide how to treat calls abandoned in the first few seconds. Most offices exclude them, but check a sample first; some are real callers who gave up.
De-duplicate repeat callers
Someone who calls three times in ten minutes is one missed caller, not three. Count unique callers per day as a second view.
Match callbacks
For each missed call, find the first outbound call or text to the same number afterwards and record the time gap.
Tag call types
Use the number dialed (listing line, sign rider, main line) and voicemail content to tag new lead, client, agent, vendor.
Example: one month for a small office
Example: a fictional eight-agent office exports October’s log. It shows 1,240 inbound calls. After removing 96 internal calls, 118 spam calls and 61 hang-ups under five seconds, 965 clean calls remain. Of those, 702 were answered live, an answer rate of 72.7% and a missed-call rate of 27.3%. Of the 263 missed calls, 151 left a voicemail. The team returned 198 missed calls within the same business day, a callback rate of 75.3%, with a median time to first callback of 2 hours 40 minutes. That leaves 65 missed calls with no return attempt at all.
Broken down by hour, the picture sharpens. Weekday answer rate between 9 a.m. and 5 p.m. is high, while Saturday 10 a.m. to 2 p.m. and weekday evenings after 6 p.m. account for most of the misses. That matches the working pattern in real estate, where evenings and weekends fit clients’ schedules[3]. The office does not need more people in the afternoon. It needs coverage on Saturday mornings and evenings.
The most important number in that example is not the answer rate. It is the 65 missed calls nobody returned. Each one is a person who called and heard nothing back.
Reading the numbers without fooling yourself
Phone metrics are easy to game by accident. A few patterns show up in almost every office that starts measuring, and each one makes the numbers look better than the caller experience.
- Forwarding to voicemail counted as answered. Some systems mark a call answered when the voicemail box picks up. Check the disposition codes before you trust the answer rate.
- Ignoring calls to personal cells. If half the listing traffic goes to agents’ own phones, a high office answer rate can hide a poor overall one.
- Averages that hide the tail. A mean callback time of one hour can include a handful of calls returned after two days. Report the median and the longest ten.
- Counting a text as a callback without a reply. A text is a good first touch, but track whether the caller was actually reached.
- Comparing different months without context. Call volume moves with the season, so a higher answer rate in a quiet month is not necessarily progress.
Also look at when people call, not only how many. Most employed people are at work on weekdays and far fewer work on weekend days[4], which is why weekend call volume is so heavy for real estate. A weekly report that shows answer rate by day and hour makes the weekend gap obvious to everyone, including agents who assume the office phones are covered.
Setting realistic targets
Avoid copying a target from another industry. Use four weeks of your own clean data as the baseline and improve it in steps.
- Split targets by period: weekday office hours, weekday evenings and weekends. One blended target hides the problem hours.
- Set a firm target for unreturned missed calls, ideally zero within one business day, before you chase a higher answer rate.
- Set time to first callback by call type: fastest for new leads and anything about a specific listing, slower for vendors.
- Raise the answer-rate target only after you change something, such as adding weekend coverage, so the target follows a real improvement.
- Review monthly. Seasonal swings are real, so compare with the same month last year once you have it.
Remember why callbacks matter so much: a large majority of Americans do not answer calls from unknown numbers[5]. A missed caller who gets a callback may not pick up either, so pair every callback with a text and a voicemail that says what the call is about.
Turn the numbers into action
Metrics are useful only when someone owns them. Give the weekly report to one person, usually the office manager or team lead, and keep it to one page: the answer rate by period, missed calls by type, median callback time and the list of unreturned calls with their owners.
- If misses cluster in a few hours, change coverage in those hours.
- If callbacks are slow, introduce a callback queue with owners and due times.
- If seller calls are being missed, prepare agents with scripts for valuation calls so the callback converts.
- If one agent’s line misses far more than others, look at routing before you look at the agent.
An SI Employee such as Ava answers every call live and logs the outcome, which lifts answer rate in the problem hours and gives you a clean record of what each caller wanted. Whatever you use, measure it the same way so you can see what changed.
Frequently asked questions
What is a good answer rate for a real estate office?
There is no single standard. Measure your own baseline for four weeks, split by weekday hours, evenings and weekends, then set targets that improve the weakest period first.
Should calls answered by an answering service count as answered?
Yes, if a person or service handled the call live and captured what the caller needed. Track who answered so you can see how much each one carries.
How do I count a caller who rang three times?
As one missed caller for that day in your unique-caller view, while the raw call view still shows three calls. Report both.
What matters more, answer rate or callback time?
Start with unreturned missed calls and callback time, because they are cheaper to fix. Then raise the answer rate in the hours where most calls are missed.
References
Zillow Research (2026). Sellers: Results from the Zillow Consumer Housing Trends Report 2026 (opens in a new tab) ↑
National Association of REALTORS® (2026). REALTORS® Adopt Technology to Save Time and Improve the Client Experience, NAR Report Finds (2026 REALTORS® Technology Report) (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). Occupational Outlook Handbook: Real Estate Brokers and Sales Agents (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). American Time Use Survey - 2025 Results (news release) (opens in a new tab) ↑
Pew Research Center (2020). Most Americans don't answer cellphone calls from unknown numbers (opens in a new tab) ↑