Key takeaways
- Build the estimate from five sources: listing and sign calls, ad and website calls, calls for agents, transaction calls and office calls.
- Multiply each driver you already know (active listings, agents, open deals) by a weekly call rate, then by 4.33 weeks a month.
- Size coverage for the busiest hour, not the monthly total. Example: one person handling six six-minute calls in an hour is already on the phone when more than a third of them ring.
- Swap the guessed rates for call-tracking and carrier data within 30 days, and rerun the estimate before each spring.
Why estimate before you count
Every phone decision a brokerage makes starts with a number it rarely has: how many calls arrive in a month, and when they land. You need it to compare a part-time receptionist with full coverage, to read an answering service quote billed per call or per minute, and to decide whether the evening calls justify a duty rotation. The best source is a logged week of real calls, which how much staff time phone calls take walks through. But plenty of offices need the number before they can log it: a new office, a team splitting off, a merger, or a broker who wants a sanity check on a vendor's figure.
The good news is that call volume follows things you already track. Listings generate calls. Advertising generates calls. Agents generate calls from lenders, title and clients looking for them. Open transactions generate calls until closing. Build the estimate from those drivers and you get a number you can defend, and a model you can correct later.
Getting it roughly right matters because the first call is often the only one. Among buyers who contacted an agent, 54% contacted only one before hiring them[2], and 67% of agent-using sellers contacted just one agent[3]. Phones are still daily work for the people taking those calls: 86% of real estate sales agents in O*NET's survey have telephone conversations every day[4].
Where brokerage calls come from
Split calls by what causes them, not by who answers. Each source has a driver you can count today and a weekly rate per unit of that driver. The rates below are starting assumptions for a resale brokerage, labelled as examples; replace them with your own as soon as you have data.
| Call source | Driver you can count | Example weekly rate | Typical callers |
|---|---|---|---|
| Listing and sign calls | Active listings | 2.5 calls per listing | Buyers asking about a home, neighbors, other agents booking showings |
| Ad, portal and website calls | Tracked numbers and campaigns | Count per number from the tracking report | Buyers and sellers responding to ads, postcards, the website |
| Calls for agents | Agents on the roster | 1.5 calls per agent | Lenders, past clients, people who lost an agent's cell number |
| Transaction calls | Deals under contract | 2 calls per open deal | Title, lenders, inspectors, clients checking on closing |
| Office and other calls | Weekdays open | 6 per weekday | Vendors, job seekers, solicitors, wrong numbers |
Transaction calls deserve their own line because they scale with production, not headcount. The typical REALTOR reported nine transaction sides in 2025, while team-based brokerage specialists reported a median of 32[5]. Two offices with the same number of agents can carry very different transaction traffic if one of them runs productive teams.
Source: National Association of REALTORS®, 2026[5]
Show the data
| Group | Median sides |
|---|---|
| Typical REALTOR | 9 sides |
| Team-based brokerage specialist | 32 sides |
Calls that ring agents' own cell phones never reach the office line, so leave them out of a main-line estimate. Do count them if you are pricing coverage for the whole brokerage, or if you plan to move agents onto a shared team number.
Example: a 25-agent office, source by source
Example. A 25-agent resale office has 30 active listings, two tracked ad numbers that together log about 40 calls a week, and 18 deals under contract. It is open 21.7 weekdays a month on average. Run each source through its rate and multiply weekly figures by 4.33 weeks.
| Source | Calculation | Calls a month |
|---|---|---|
| Listing and sign calls | 30 listings × 2.5 × 4.33 | 325 |
| Ad, portal and website calls | 40 a week × 4.33 | 173 |
| Calls for agents | 25 agents × 1.5 × 4.33 | 162 |
| Transaction calls | 18 deals × 2 × 4.33 | 156 |
| Office and other calls | 6 a weekday × 21.7 | 130 |
| Total | All five sources | 946 |
Example. That is about 946 calls a month, or roughly 38 per agent. Next, split the total by time of day. People call when they are free, and for most of them that is outside working hours: 81% of employed people work on an average weekday, against 30% on an average weekend day[6]. Agents' own schedules often include evenings and weekends for the same reason[7]. Assume a quarter of calls arrive after 6 pm or at weekends until your data says otherwise.
- Example business hours: 946 × 75% ≈ 710 calls a month, about 33 a weekday.
- Example evenings and weekends: about 236 calls a month, roughly 55 a week.
- Example calls per agent: 946 ÷ 25 ≈ 38 a month, a useful figure when someone quotes coverage per agent seat.
Compare your result with a known case before trusting it. The 40-agent front desk example works from about 60 main-line calls a weekday plus roughly 90 a week in the evenings and at weekends, with twice the listings. If your estimate is wildly off that ratio, recheck the listing rate first; it is the input most people overstate or forget.
Seasons, launches and rate moves
A monthly average hides the months that hurt. Realtor.com's analysis of 2018 to 2025 data found new sellers historically climbed to nearly 1.4 times their start-of-year level by the end of June[1], and listings in its best spring week drew 16.7% more views per listing than a typical week[1]. More listings and more views mean more listing and ad calls, while calls for agents and office calls barely move. Seller activity also tends to keep climbing until roughly November[1], so the slow stretch is short.
≈850
calls in a slow winter month (listing and ad calls × 0.8)
≈950
calls in a typical month
≈1,120
calls in a spring peak month (listing and ad calls × 1.35)
≈6
calls in the busiest single hour of the week
Example figures from the worked estimate above. Only listing and ad calls are scaled; the other sources are held flat.
Then add the spikes you cause yourself. A new listing's first weekend, an open house, a postcard drop and a price reduction all pull calls into a few days. So does a drop in mortgage rates, which can flood the phones with buyers who were waiting; preparing for a rate-drop surge covers that case. Keep a calendar of launches next to the estimate and add a line for each one rather than raising the whole month.
The busy hour decides the staffing
Monthly totals price a service. The busy hour tells you whether one person can keep up. Calls do not arrive evenly: Monday mornings, the hour after lunch and late Saturday morning usually run hot, and peak call times for real estate offices explains how to find yours. Example assumption until you have a log: the heaviest weekday carries a quarter more than average, and its busiest hour takes about 15% of that day's calls.
Example. For the 25-agent office, the heaviest weekday brings about 41 calls (33 × 1.25), and the busiest hour about six of them. If each call takes six minutes including the note afterwards, the Erlang B formula that phone planners use says one person will already be on a call when about 37% of those six callers ring. With two people answering, that drops to about 10%. The office is nowhere near full for the day, yet a third of its peak-hour callers meet a busy line, voicemail or a ring group that rolls on.
Rule of thumb from the example: if your busy hour holds more than about four calls of five minutes or longer, one person cannot cover it without callers waiting or rolling to voicemail. Plan an overflow path for that hour, not another full-time hire.
Overflow can be a second person on the ring group, an agent on floor duty, or forwarding unanswered calls after a set number of rings. Whatever you choose, measure what happens to the overflow: how to measure answer rate and missed-call rate sets up the two numbers that tell you whether the plan works.
Turn the estimate into desk hours
With calls per weekday in hand, the staff-time model converts them into monthly hours and cost. Price the time at the full cost of the person answering, wages plus benefits: private employers paid an average of $36.73 per hour worked in total compensation for office and administrative support occupations in June 2026[8]. The handled share is the part of calls the desk finishes without passing them on; the rest still reach agents.
| Input | Example |
|---|---|
| Weekday calls from your estimate | 33 |
| Minutes per call, including the note | 6 |
| Share the desk finishes without an agent | 70% |
| Hourly cost of desk time (wages and benefits) | $36.73 |
| Calls a month | 716 |
| Staff hours on these calls | 72 |
| Hours handed off | 50 |
| Monthly value of that staff time | $1,841 |
How it is worked out: Calls a day × 21.7 working days × minutes per call ÷ 60 × share handled = hours; hours × hourly cost.
Assumptions
- Hourly cost is the June 2026 average total compensation for office and administrative support occupations in private industry.[8]
- Six minutes per call and a 70% handled share are example inputs; use your own once calls are logged.
- Weekday calls are the business-hours share (75%) of the worked estimate, spread over 21.7 weekdays.
Example. Change the inputs to match your office. The result covers weekday business hours only.
Example. The answer is about 50 desk hours a month, roughly $1,840 at that hourly cost. Full-time employees averaged 8.1 hours on days they worked in 2025[6], so that is a little over six working days a month spent on calls. It is not a full-time job, which is exactly the point the busy-hour calculation makes: the problem is timing, not total hours. The 236 evening and weekend calls sit on top and need their own answer, whether that is an agent rotation, forwarding or a service.
This is also the point where the estimate meets a buying decision. If you are comparing a part-time receptionist with full phone coverage, run the same numbers for the hours each option covers and look at which calls fall outside them.
Four ways the estimate goes wrong
- Counting callers instead of calls. A buyer who rings three times about the same house is three calls for the person answering, even if it is one lead in the CRM.
- Using the phone bill's answered calls only. Calls that rang out, hit a full voicemail box or hung up during the greeting are demand you failed to see.
- Forgetting the agents' share. Lenders and title officers who cannot reach an agent's cell call the office, and they call more when deals are close to closing.
- Averaging the year. A spring month can run well above a winter one, so a staffing plan built on the average is short-handed exactly when listings peak.
Each of these makes the office look quieter than it is. The fix is the same for all four: count calls at the carrier, by hour, for a full month, and keep the source tags.
Replace the estimate with real numbers
Treat the estimate as a first draft that expires in a month. Every rate in it is a guess you can now measure.
Pull the carrier records
Your phone provider's call detail records show every inbound call to the main line with date, time and duration, including the ones nobody answered. Count a full month.
Put a tracked number on each source
Give sign riders, portals, postcards and the website their own numbers so the listing and ad rates come from data. Call tracking numbers for listings and ads covers the setup.
Tag call reasons for one week
Have whoever answers mark each call as listing, ad, agent, transaction or office. That gives you the real mix behind the total.
Count what never connected
Missed and abandoned calls belong in the volume. Callers who gave up still wanted something, and some will not try again.
Rerun the model
Replace each example rate with your measured one, rebuild the low, typical and peak months, and recheck the busy hour.
Forwarding helps here too. When unanswered or after-hours calls forward to an SI Employee, each one arrives in your inbox with a recording, transcript and summary, so the reason for every overflow call is on record instead of lost in a voicemail box. The real estate team lead response example shows how that kind of coverage is modelled for a team. And when the count shows callbacks piling up after busy hours, a callback queue keeps them in order.
Frequently asked questions
How many calls does a real estate brokerage get per agent?
There is no published standard, because it depends on listings, advertising and production. In the worked example, a 25-agent resale office with 30 listings comes to about 946 main-line calls a month, roughly 38 per agent. Build your own figure from the five sources and check it against your carrier records.
Should calls to agents' cell phones count?
Not for a main-line estimate, because the office never sees them. Count them when you are pricing coverage for the whole brokerage or moving agents onto a shared team number, since those calls will then reach the line you are staffing.
Why plan for the busiest hour instead of the daily total?
Because calls arrive in clusters. In the example, six six-minute calls in one hour keep a single person busy when more than a third of them ring, even though the day as a whole is light. Size the overflow path for that hour.
How often should the estimate be redone?
Rerun it once with real data a month after you start, then before each spring market. New sellers have historically reached nearly 1.4 times their start-of-year level by the end of June[1], so last winter's numbers will undercount the spring.
Do outbound calls belong in the volume?
Keep them separate. Inbound volume decides how many people must be free to answer. Outbound callbacks add workload to the same people, so include them when you convert calls into desk hours.
References
Realtor.com Economic Research (2026). The Best Time To List a Home: The Week of April 12-18 (2026 analysis) (opens in a new tab) ↑
Zillow Research (2026). Buyers: Results from the Consumer Housing Trends Report 2026 (opens in a new tab) ↑
Zillow Research (2026). Sellers: Results from the Zillow Consumer Housing Trends Report 2026 (opens in a new tab) ↑
O*NET OnLine, U.S. Department of Labor (2026). 41-9022.00 - Real Estate Sales Agents (Details: Work Context) (opens in a new tab) ↑
National Association of REALTORS® (2026). Experienced REALTORS® Anchor the Industry as Housing Affordability Remains Top Hurdle (2026 Member Profile) (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2026). American Time Use Survey - 2025 Results (news release) (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 (2026). Employer Costs for Employee Compensation, Table 4: Private industry workers by occupational and industry group (June 2026) (opens in a new tab) ↑