From Crawlspace to Five Stars: A 5-Step Review System for Plumbing Companies Whose Techs Hate Asking

Your tech just spent four hours in a crawlspace replacing a failed water heater. The homeowner had been without hot water for two days, and now she's thanking him at the door like he saved the house. He nods, packs the truck, and drives to the next call. No review gets requested, written, or posted. Run six jobs a day across five trucks, and that's roughly 150 of those moments every week — nearly all of them evaporating before anyone thinks to capture one.

The leak: jobs that convert, reviews that don't

Most plumbing companies are sitting on a review deficit that has nothing to do with the quality of their work. About 88% of consumers trust online reviews as much as personal recommendations, which means your Google profile is doing the same job your best word-of-mouth customer does — except at scale, for every stranger who types "plumber near me" at 9 p.m. with water on the floor. The math on ratings isn't subtle either: one additional star in average rating correlates with roughly a 12% lift in conversion. A shop completing 30 jobs a week and collecting three reviews a month isn't unlucky. It has a leak: satisfied customers walking away un-asked, every single day, while a competitor with worse work and better process stacks social proof.

Why the ask never happens

It isn't laziness. Techs are tradespeople, and asking for a favor at the door feels like begging — the good ones are already mentally on the next job before they've closed the truck doors. The office intends to follow up "when things slow down," which turns into a Friday batch of texts sent three days after the goodwill peaked, or into nothing at all. And nobody owns the number. Revenue has an owner. Callbacks have an owner. Warranty claims have an owner. Review velocity usually belongs to everyone, which means it belongs to no one. So the single best moment to ask — the hour after the hot water comes back on — reliably passes in silence, hundreds of times a month.

The 5-step system

You can fix most of this with process before you spend a dollar on software. Here's the system we'd run in any plumbing operation, starting tomorrow.

1. Trigger the ask off "job closed," not off a calendar. The moment your tech marks a job complete in your field-service software is the moment the customer's relief is at its peak. That event — not a Tuesday batch, not a monthly newsletter — should start the clock on the review request. If your software can't fire a notification on job completion, have dispatch keep a running list and work it same-day. The principle is simple: the ask should arrive while the customer is still telling their spouse how fast you got there.

2. Turn the tech's part into a handoff, not a pitch. Don't ask techs to sell. Give them one scripted sentence for the doorstep: "You'll get a text from us in a little bit — it has a link if you'd ever want to leave us a review." That's it. No pressure, no awkwardness, no favor being requested face-to-face. It primes the customer so the text that follows feels expected instead of spammy, and it takes exactly four seconds. This step matters more than it looks: a text that arrives cold gets treated like spam, while a text the customer was told to expect gets opened and acted on. Put the sentence in onboarding, laminate it on the clipboard, and move on.

3. Send one text with one link within the hour. From a local number, not a short code. Name the tech and the job: "Thanks for having Marcus out today for the water heater replacement." Then one link, straight to your Google review form — not to your homepage, not to a "review us on your platform of choice" menu. Every additional tap loses people. Specificity matters more than polish here: a text that names the tech and the job reads like it came from a real shop, and it also nudges the customer toward writing the kind of detailed review that persuades the next reader. "Marcus was on time and explained the options on our water heater" does far more work on your profile than a bare five stars. One message, one destination, sent while the experience is still warm.

4. Route by job type, because not all jobs carry equal emotion. A $180 drain clearing and an emergency repipe after a burst line are different events in the customer's life. Emergency work — sewage backups, burst pipes, no hot water in January — carries the most relief and produces the most vivid, persuasive reviews. Flag those job types in your system and make sure they never slip through the cracks. If you only have the discipline to run this system on part of your volume, run it on the emergencies first.

5. Follow up once at 48 hours, then stop — and put the number on the wall. One reminder is a courtesy; three is harassment that earns you the wrong kind of review. After the single follow-up, let it go. Then make review velocity a tracked number with an actual owner: reviews per completed job, weekly, visible to the whole shop, ideally broken out per tech. Give the dashboard to your office manager or whoever runs dispatch — someone who will notice within a week when the texts stop going out, not within a quarter. Once techs see their names next to a number, the doorstep sentence stops getting skipped, and the shop's monthly review count becomes something people compete over instead of something nobody remembers to check. What gets counted gets done — that's as true in a plumbing shop as anywhere else.

Run those five steps manually for 30 days and you will almost certainly see the needle move. The constraint you'll hit isn't whether the system works — it's whether anyone keeps doing it in week six, when the office gets busy.

The proof

Case study: AMG Transport Co., a transportation and logistics operator using Tykon's system, went from 5–6 new reviews a month to 5–6 a day once post-job review requests were automated. Different trade, same mechanics: the jobs were already producing satisfied customers every single day. The reviews weren't missing because customers were unwilling — they were missing because the asks weren't going out. Once the ask became automatic, tied to the completed job instead of someone's memory, the review count simply caught up to the work that was already being done.

Where automation picks this up

Everything above runs on paper and discipline. The failure mode isn't the design — it's entropy. The dispatcher gets slammed, the texts slip to end of day, the 48-hour follow-up stops happening, and by month two the system exists only in the training binder. That's the specific gap Tykon's agent, James, closes: it triggers off job completion automatically, sends the text and the single reminder, and then rolls your happiest customers into a referral prompt afterward — every job, every day, without anyone at the office touching it. If you want a gut-check before talking to anyone: pull your last 30 days of completed jobs and count them against new Google reviews in the same window. If the two numbers embarrass each other, the leak is real — and it's worth 15 minutes to see how the automated version runs.