Net Promoter Score and post-job surveys

Why a post-job survey is worth the effort

Reviews tell you what a small share of customers chose to say in public. A short survey tells you what most customers think, including the quiet ones who will never write a review and simply call someone else next time. For home service companies built on repeat work and referrals, that quiet group matters.

Repeat demand is real. In Angi's 2025 survey, the average number of home projects completed per household rose from 9 to 10^[3]. A household that does ten projects a year is a customer you can serve again and again, or lose after one bad visit. A survey gives you a chance to find and fix the bad visit before it becomes a lost customer or a public complaint.

How NPS works

Bain & Company publishes the Net Promoter Score method. The core question is 'How likely are you to recommend us to a friend or colleague?', answered on a zero-to-ten scale^[1]. Customers who answer 9 or 10 are promoters, 7 or 8 are passives, and 0 to 6 are detractors. The score is the percentage of promoters minus the percentage of detractors^[1], so it ranges from minus 100 to plus 100.

Example: a fictional HVAC company receives 120 responses in a month. Seventy-eight customers answer 9 or 10, which is 65%. Twenty-four answer 7 or 8. Eighteen answer 6 or below, which is 15%. The NPS is 65 minus 15, or 50. Passives count in the total but not in the subtraction.

Designing a survey people will answer

Short wins. Every extra question lowers the number of people who finish. For most home service companies, three questions are plenty.

1. How likely are you to recommend [company] to a friend or neighbor? (0 to 10)
2. What is the main reason for your score? (open text)
3. Optional: Did [technician first name] arrive on time and leave the work area clean? (yes or no)

• Send it by text from the number the customer already knows, within 24 hours of the job.
• Use a mobile-friendly link that opens to the first question, no login.
• Send one survey per job, and for recurring services no more than once a quarter per customer.
• Say how long it takes: 'Two quick questions about today's visit.'
• Do not include a review request inside the survey itself. Send review requests separately, to everyone.

Keep surveys and review requests separate

Many survey tools offer a feature that sends promoters to Google and detractors to a private form. It looks clever and breaks the rules. Google's Maps content policy says merchants may not discourage negative reviews or selectively solicit positive reviews from customers^[2]. FTC guidance adds that asking only customers you expect to be happy would be misleading if it skews how favorable your reviews look^[4].

The safe setup is two independent messages. Every customer gets the same review request after the job. Every customer, or a regular sample, gets the survey. The survey score decides who gets a follow-up call, never who gets asked for a review. Readers notice the difference too: Pew found that 82% of adults at least sometimes read reviews before a first purchase^[5], and a profile with only glowing reviews invites suspicion.

Closing the loop

A survey without follow-up teaches customers that feedback goes nowhere. Set rules for each group and track that they are followed.

1. 01

   Detractors (0 to 6): call within one business day

   A manager, not the technician, calls. Listen first, ask what would make it right, agree a next step and confirm it in writing. Log the root cause.

2. 02

   Passives (7 or 8): read every comment

   Passives often name a small fix: late arrival, unclear price, a missed detail. Reply by text to thank them and, where useful, ask one follow-up question.

3. 03

   Promoters (9 or 10): thank and invite referrals

   Send a short thank-you. If you have a referral program, mention it here. They have already received the same review request as everyone else.

4. 04

   Weekly: review root causes

   In the weekly team meeting, read three comments aloud: one detractor, one passive, one promoter. Agree one action.

5. 05

   Monthly: track the trend

   Report NPS, response count and detractor callbacks completed by service line and technician.

Reading the results without fooling yourself

• Watch the number of responses. A technician's NPS from eight responses swings wildly; wait for at least 30 before drawing conclusions about one person.
• Read comments before scores. 'Great tech, but the office never called back' is two different findings.
• Look for patterns by job type. Installs and emergencies often score differently from maintenance visits.
• Do not reward or punish technicians on NPS alone. Use it for coaching alongside callbacks, reviews and quality checks.
• Never ask customers to raise their score or tell them what score 'counts'. That poisons the data.

Survey results also feed other work. Put the top three complaint themes on your customer journey map, use detractor calls as the start of your complaint resolution process, and add the monthly NPS line to your marketing review.

Getting enough responses

A survey that only a handful of customers answer tells you little. Response rates depend more on the details of the request than on the questions themselves.

• Mention the survey at the end of the visit: 'You'll get a two-question text from us tomorrow. We read every answer.'
• Send during the day, not late at night, and only once per job.
• Put the first question in the text itself where your tool allows it, so the customer can reply with a number.
• Follow up on comments quickly. Customers who see that someone acted on their feedback are more likely to answer next time.
• Do not offer prizes or discounts for completing the survey if the survey is in any way linked to reviews. Keeping it simple avoids the problem.

What you need to run it

You do not need specialist software to start. Many field service platforms include a basic survey that triggers when a job is completed. If yours does not, a simple form tool and a text message from your office number will do for the first few months. What matters is that responses land somewhere a manager sees every day, that each response is linked to the job and technician, and that detractor responses create a task with an owner and a due date. Start small, prove the routine works, and only then look at dedicated tools.

Example: what a quarter of surveys can change

Example: a fictional lawn care company surveys each customer once per quarter. In the first quarter, the most common detractor comment is 'I never know when the crew is coming.' The owner adds a day-before text and an on-the-way text. In the second quarter, that comment almost disappears, and the top passive comment becomes 'missed a strip along the fence.' The crew leads add a fence-line check to the end-of-visit routine.

Neither change was expensive. Both came from asking customers two short questions and reading the answers. That is the real value of NPS: not the score, but a steady flow of specific, fixable problems in the customers' own words.