A home-repair company goes digital, part 5: make feedback the lifeline — the service, feedback, improve, re-discovery loop

This series follows a home-repair company in Chonburi, step by step, into digital. The story comes from our real implementation experience in the home-services industry — if you run a small service company living off repeat customers, you should see your own vans in every chapter. Each chapter explains: what was built, what was deliberately not built, and what evidence earned the right to move on.

The end of one repair is the start of the next service

For a home-repair company, the end of a service usually means: the technician collects the money and leaves, the company loses contact, until the customer’s home has a problem again. During that gap, the customer is “out of reach.”

The PWA changes more than the “entry” — it lets the company stay in the customer’s life during that gap. The method is one word: ask.

The four steps of the loop: service, feedback, improve, re-discovery

Step one: service. The technician completes the repair — the most familiar part for the company. The only change: before leaving, tell the customer “when you get home, tap the icon on your phone and you can give us feedback — next time you have a problem, come back through that same icon.”

Step two: feedback. After service, the company sends a notification through the PWA inviting the customer to rate this service: was the technician on time, was the problem solved, was the quote clear, will you come back to us. Feedback is tied to the specific technician — because only by knowing “which technician earned this good review, who caused this complaint” does feedback have improvement value.

Step three: improve. The owner reads the feedback every week. Not to enjoy or feel bad about scores, but to see patterns: a technician keeps getting “quoted low then charged more,” a certain job keeps getting “not fully fixed,” AC-season appointments always can’t fit in. The feedback data tells the company what to fix — something impressions could never reveal.

Step four: re-discovery. The customer was served well, received a feedback invitation, and has the company icon on the home screen. Next time the home has a problem, they don’t scroll chats — they tap the icon and re-book directly. Returning customers shift from “by luck” to “through a channel.”

Once the loop turns, the company’s growth logic changes: no longer “every order pulls a new customer,” but “serve one well, keep one, make them come back every year.”

How technicians view “feedback will be tied to me”

This is where it can flip the most. If technicians feel feedback is something the boss uses to dock pay, they will: first, discourage customers from giving it; second, grab the easy high-score jobs and push the bad ones onto others. The loop dies on the spot.

So the company set a few rules in advance:

• Feedback’s first use is improving service, not grading pay. What technicians see is “which jobs keep getting complained about, how to avoid them next time” — not “where does your score rank”;
• Using feedback for evaluation is a later independent step. Only after enough data has accumulated, rules are clear, and dispute handling is defined does the company discuss bringing feedback into evaluation — not promised this phase;
• Technicians get a say too. On a low score, the company talks to the technician about what happened on site first, instead of convicting them. Technicians cooperate because they know feedback is helping them, not watching them.

The feedback loop survives not because the system is smart, but because the technicians believe it’s a helper, not a shackle.

What this phase was deliberately not

• Not a technician-evaluation system. Feedback first improves service; evaluation is a later independent step, and this phase builds no score ranking;
• Not a full CRM or membership system. No full customer profiles, no points, no automated marketing — the PWA remembering “who’s a returning customer” is enough;
• Not a dispatch scheduling system. The loop doesn’t solve who goes where; it solves whether the customer comes back;
• Not a native app. A home-screen icon plus notifications — the PWA already covers this scenario.

How the loop is judged a success

After the loop has run a while, the company looks at three things:

• Has feedback reached scale — of the notifications sent, how many customers actually answered;
• Has feedback driven improvement — did the complained-about patterns really get better, instead of stopping at “feedback received”;
• Has re-discovery increased — is the share of re-books through the PWA rising steadily.

Part 6 looks back at both phases together: how far the wasted-visit rate fell, how repeat-customer return changed, why each step was decided that way, and how this case held the “no large systems, no app” line.

---

Boundary matters: the PWA feedback notifications and loop in this phase are the company’s phase-two independent project; using feedback for technician evaluation is a later independent step, and this phase builds no scoring system. Dispatch scheduling, online payment, customer-account systems, and full CRM are all outside this case’s scope and do not become part of a standard website package by appearing in it.