Key takeaways
- Track seven numbers each week: booking rate, answer rate, handle time, cancellation rate, add-on rate, membership offer rate and QA score.
- Define each one in writing, especially what counts as a bookable call, or the numbers will be argued rather than used.
- Set targets from your own last 90 days, then move them gradually; borrowed benchmarks rarely fit your call mix.
- CSR turnover is high, with about 289,500 openings projected each year nationally[1], so use the scorecard for coaching, not punishment.
Why a weekly scorecard
A CSR team that only hears about results when something goes wrong learns to avoid blame, not to book more jobs. A short weekly scorecard gives each person a clear picture of how they are doing and gives the manager a reason to coach every week, not only after a complaint.
It also helps keep people. Customer service is a high-turnover role: BLS counts about 2.7 million customer service representative jobs and projects about 289,500 openings each year, all expected to come from replacing workers who move to other jobs or leave the labor force[1].
Source: U.S. Bureau of Labor Statistics, 2025-2035[1]
Show the data
| Item | Value |
|---|---|
| Jobs, 2025 | 2,700,000 |
| Openings each year | 289,500 |
Building equipment contractors alone employed 21,580 customer service representatives in May 2025[2]. Every CSR who leaves takes their training with them, and private employers' compensation costs averaged $46.89 per hour worked in June 2026[3], so the hours spent hiring and training a replacement are expensive. A scorecard that shows progress and gives fair feedback is one of the cheapest retention tools a manager has.
The seven numbers
| Number | Definition | How to calculate | Example target |
|---|---|---|---|
| Booking rate | Bookable calls that ended with a booked job or estimate | Booked calls / bookable calls | Your 90-day average plus a few points |
| Answer rate | Inbound calls answered by a person | Answered calls / inbound calls | Set by hour of day |
| Handle time | Average talk plus wrap-up time per call | Total handle minutes / calls handled | Track for outliers only |
| Cancellation rate | Jobs booked by the CSR that cancelled or no-showed | Cancelled jobs / jobs booked | Your 90-day average or lower |
| Add-on rate | Booked jobs where the CSR added a related service | Jobs with an add-on / jobs booked | Agree per job type |
| Membership offer rate | Eligible callers offered a maintenance membership | Offers made / eligible calls | Every eligible call |
| QA score | Average score from the call quality rubric | Sum of scores / calls scored | Your rubric's pass mark |
Booking rate: define the denominator
Booking rate is the most important number and the most argued. The arguments are almost always about the denominator. Decide in writing which calls count as bookable.
- Count: new and existing customers calling about a service you offer, in your area.
- Exclude: wrong numbers, spam, job seekers, vendors, calls about existing appointments and out-of-area callers.
- Decide how to handle price shoppers who hang up. Most managers count them, because a good CSR can book some of them.
- Tag calls consistently. Call tracking or phone system tags make this possible; our guide to choosing a phone system covers reporting.
Once the definition is fixed, keep it fixed. Changing it every month makes the trend meaningless.
Answer rate and handle time
Answer rate shows whether callers reach a person. Look at it by hour, not just as a weekly total, because a strong average hides the lunch hour or the Monday morning rush. If it drops at the same hours every week, the fix is a schedule or routing change, not a CSR problem.
Handle time is useful for spotting outliers, not for squeezing calls. A CSR whose calls are much shorter than everyone else's may be skipping questions; one whose calls are much longer may need help with a script. Listen to a few before drawing conclusions.
Cancellations, add-ons and memberships
Cancellation rate by CSR shows who is booking jobs that stick. A high rate can mean missing confirmation steps, unclear windows or bookings made without the decision-maker. It can also mean nothing more than a CSR who handles more online leads, so compare like with like.
Add-on rate measures whether CSRs mention related services the customer would want, such as a second system's tune-up or a whole-home surge protector with a panel visit. Keep it honest: the add-on must fit the caller's situation.
Membership offer rate measures offers, not sales. CSRs control whether they offer; customers decide whether to buy. Tracking the offer keeps the habit alive without pushing CSRs into hard selling.
A membership offer that isn't pushy
CSRs skip membership offers when they feel like a sales pitch tacked onto a service call. Give them a short, honest line that fits the conversation, and use it after the booking is confirmed, never before.
"You're all set for Thursday between 8 and 10. One thing worth knowing: we have a maintenance plan that includes your yearly tune-up and priority scheduling when the weather turns. Would you like me to text you the details, or have the technician go over it on Thursday?"
The question gives the customer an easy way to say "send me the details," which counts as an offer made and moves the conversation to the technician, who can explain it in person. Track the offer rate weekly and the sign-up rate monthly; the second depends on the plan and the technician as much as the CSR.
When a number drops
A falling number is a question, not a verdict. Before the coaching conversation, check three things in order.
- Did the call mix change? A new marketing campaign, a heat wave or a run of existing-customer calls can move booking and handle time without any change in the CSR's skill.
- Did something outside the CSR's control change? A full schedule, a new diagnostic fee or a phone outage shows up first in the CSR's numbers.
- What do the calls sound like? Listen to two or three calls from the week before deciding anything.
Only when the first two are ruled out is it a coaching issue, and then the coaching point should be one specific behavior the CSR can practice.
QA score: a short rubric
Score a small sample of calls against the same rubric every week. A simple rubric has eight to ten items, each scored yes or no:
- Greeted with the company name and their own name.
- Confirmed the service address and a callback number early.
- Asked safety questions where relevant: gas, sparks, carbon monoxide, water near electrics.
- Used the trade's question sheet to choose the right job type.
- Offered two specific windows.
- Explained the diagnostic or trip policy clearly.
- Offered a membership or add-on where it fit.
- Confirmed the booking and the confirmation text.
- Stayed calm and polite throughout.
- Wrote clear notes for the technician.
Setting targets
Measure 90 days first
Pull the last three months for each number, by CSR and by call type, before setting any target.
Set team targets from your own data
Start with your current average and set a modest improvement. Borrowed industry numbers rarely match your marketing mix.
Set individual goals from each CSR's baseline
A new CSR improving fast matters as much as a veteran holding steady.
Review targets quarterly
Seasonality changes the call mix, so a summer target may not fit winter.
Tie targets to coaching
Every target that is missed gets a specific coaching action, not just a note.
Running the weekly review
- Send each CSR their own numbers and the team totals by Monday morning.
- Meet each CSR for fifteen minutes and listen to one good call and one call to improve.
- Agree one thing to work on for the week, written down.
- Note anything outside the CSR's control, such as a full schedule or a phone outage.
- Share team wins in the team meeting.
Context matters: when the board is full, booking rate falls for reasons no script can fix; see what to say when the schedule is full and technician capacity planning.
Example: one scorecard, one coaching point
Example: a fictional HVAC company's newest CSR has a lower booking rate than the team average, but her QA scores are strong on safety and notes. Listening to two unbooked calls, the manager hears her ask "When would you like us to come out?" instead of offering two windows. Callers hesitate and say they'll call back. The coaching point for the week is a single sentence: "I can have someone there tomorrow between 8 and 10, or Thursday between 1 and 3." Two weeks later her booking rate has moved toward the team average, and her QA score is unchanged.
Pitfalls
- Too many numbers. Seven is plenty; more becomes noise.
- Targets without definitions, which turn reviews into arguments.
- Ranking CSRs publicly, which encourages gaming, such as tagging hard calls as unbookable.
- Ignoring callbacks. Outbound calls matter too, and only 19% of adults generally answer calls from unknown numbers[4], so track whether callbacks reach customers.
- Scoring only complaint calls, which makes QA feel like punishment.
- Leaving out commercial calls, which need their own fields and checks, as covered in commercial service call intake.
Frequently asked questions
What is a good booking rate for a home service CSR?
It depends on your call mix and marketing. Measure your own rate on bookable calls for 90 days, then aim to raise it step by step. Compare CSRs with each other on the same call types, not with numbers from another company.
Should handle time have a target?
Track it, but be careful targeting it. Pushing handle time down tends to cut questions that prevent wrong bookings and add-ons. Use it to spot outliers, such as very short calls that never book.
How many calls should we score for QA each week?
A small, regular sample works better than a big monthly review. Many managers score three to five calls per CSR each week, chosen at random plus any flagged complaints.
Should CSRs see each other's numbers?
Sharing team totals is useful; ranking individuals publicly often backfires. Review individual numbers one to one, and celebrate improvements in the team meeting.
References
U.S. Bureau of Labor Statistics (2025-2035). Customer Service Representatives: Occupational Outlook Handbook (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2025). Occupational Employment and Wage Statistics, May 2025, series OEUN000000023820043405101 (Customer service representatives in Building Equipment Contractors) (opens in a new tab) ↑
U.S. Bureau of Labor Statistics (2026). Employer Costs for Employee Compensation — June 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) ↑