The expensive silence after every job
Every completed job produces one of three outcomes: a customer who would rave about you, a customer who would warn others away, and a customer who felt fine but unremarkable. Most local service businesses never find out which — until the happy customer forgets to leave a review and the angry one posts theirs publicly, star rating and story included.
That asymmetry is what a closed-loop feedback system fixes. It is an operational instrument: a short survey delivered at the right moment, with routing rules that decide what happens next. Detractors get a human callback before they vent publicly; promoters get sequenced asks; every verbatim response feeds your marketing, sales training, and quarterly leadership review.
This is deliberately not a review-velocity play. Review generation is one downstream use; the spine of the system is the survey and the closed operational loop around it — because the businesses that win are not the ones with the most feedback, but the ones that act on feedback fastest.
The instrument: short enough that people actually finish it
Your survey competes with everything else on a customer's phone. Keep it under sixty seconds; three components do the work:
- One rating question. The classic is the 0–10 likelihood-to-recommend scale — not because the number is magical, but because it gives a clean trend line and a clear routing threshold. A 1–5 satisfaction rating works too — pick one and never change it, or your trends break.
- One open-ended question. “What one thing could we have done better?” for everyone, or “What did we do well?” for promoters. The verbatim text is where your marketing copy lives — more on that below.
- Optional tag-on. Permission to use their comments in marketing (“May we quote you?”). One checkbox, captured while the goodwill is fresh.
Resist adding questions about each department and crew member — every added question cuts completions, and completions are the fuel of the system.
Timing and channel: deliver it when the experience is freshest
Timing matters more than tooling. Send the survey when the job's outcome has settled but before memory fades — typically 24 to 72 hours after completion. Same-day surveys catch customers still reacting to punctuality; a week later, the details that make verbatim responses useful have faded. For larger projects like remodels, wait for a natural milestone — the walkthrough, punch-list signoff — not the last day on site.
Channel mix, in order of effectiveness for most local service audiences:
- Text message first. Short link, no login, no app. This is usually the highest-completion channel for service customers.
- Email as the backstop for numbers you cannot text and for customers who ignore the first ask.
- The in-person seed. The tech or project manager says one sentence at closeout: “You'll get a quick text from us — it takes under a minute, and I read every response.” Expectations set face-to-face improve response quality — the survey now feels like a conversation continuation, not marketing spam.
Never survey a customer mid-complaint: with an open ticket, the survey reads as tone-deaf and poisons the data. Suppress active complaints from the send list automatically.
The closed loop: routing is where the money is
A score that nobody acts on is worse than no score at all — it creates data theater. Every response needs a route:
- Detractors (0–6): Instant alert to a named human — owner, service manager, or recovery owner — with a mandate to call within hours, not days. The goal is to hear the complaint before it becomes a review or a chargeback. Track whether the save worked; “closed” means the customer confirmed resolution, not that you left a voicemail.
- Passives (7–8): Route to operations, not marketing. Passives liked the work but felt nothing special — often a communication, cleanliness, or follow-up gap. A quick check-in call converts many into future promoters, and their recurring themes are your cheapest operational improvements.
- Promoters (9–10): This is where review velocity connects — as a downstream step, not the point. Send the review ask within a day of the survey while enthusiasm holds. Referral asks come later, after the review posts or a few weeks pass. Asking for both in the same breath reads transactional and suppresses both.
Mining verbatim responses for copy you could never write yourself
Customers describe your value in words your copywriter will not invent. A remodeling client writes “they treated my house like it was their own.” A plumbing customer writes “no surprise on the invoice.” Those sentences belong on your homepage, in your ads, and in your sales team's mouth.
Build a light tagging habit: once a month, skim responses and tag recurring themes — communication, cleanliness, punctuality, pricing clarity, cleanup. Praise themes become headline candidates; complaint themes become operational fixes and objection-handling copy. “No surprise on the invoice” beats any generic “quality workmanship” ad claim, because it is real customer language. Get permission via the survey checkbox, and attribute first name + neighborhood, never full details.
Quarterly reporting: make the trend impossible to ignore
Once a quarter, compress the whole system into a one-page leadership report:
- Score trend quarter over quarter — direction matters more than the absolute number.
- Response and completion rates by channel, so you know if your delivery is decaying.
- Save rate — detractors recovered versus total detractors — which is your service-recovery team's batting average.
- Loop throughput — what percentage of promoters actually received a review ask, and what percentage posted one.
- Theme shifts — which complaint tags are rising or falling, and which praise tags are becoming signature strengths worth marketing.
Compare survey themes against public review themes each quarter. When they diverge — healthy survey scores but reviews turning negative, or vice versa — something in routing or timing is off. That comparison alone justifies the system: it catches blind spots before they reach revenue.
The playbook in steps
- Draft the three-part survey: one 0–10 rating question, one verbatim question, one quote-permission checkbox. Keep it under sixty seconds.
- Pick the delivery rails: SMS first, email backstop, wired into your CRM or field-service platform so completion status updates automatically.
- Set the timing rule: 24–72 hours post-completion as the default; milestone-based timing for long projects; automatic suppression of open-complaint customers.
- Assign owners to each route: a named recovery owner for detractors with an hours-not-days callback mandate, an ops owner for passives, a marketing owner for promoter asks.
- Sequence the promoter flow: review ask within a day, referral ask later, never simultaneous.
- Start the verbatim habit: monthly theme tagging, feeding copy, objection handling, and operational fixes.
- Ship the one-page quarterly report and review it in a leadership meeting like a P&L line — same chair, same cadence, every quarter.
Pitfalls that quietly kill these systems
The failures are predictable. Surveys sent too early or too late return noise. Incentivizing responses skews toward happy customers and poisons the trend line. Routing detractors to a generic inbox instead of a named human guarantees the public vent you were trying to prevent. And treating the survey as a one-and-done project produces a year of stale data and a leadership team that stops believing in it.
The fix for all of these is the same: treat the feedback loop as an operating rhythm, not a campaign. The businesses that do see reviews improve, referrals compound, and marketing start speaking the customer's own language. That is what it means to architect a sales engine instead of a brochure.