Customer reviews sorted to find what is behind the low stars
Two years of customer reviews sorted by theme to find the complaint behind your low ratings and when it started. For service businesses in Northeast Arkansas.
Demonstration using synthetic data
Every record, name, address, and log entry on this page is invented for the demonstration. No client, client system, or production record is involved, and no business result is claimed.
- Author
- Tanner Rusher
- Published
- Written for
- Trades and service businesses
Who this is for
Owners of heating and air, plumbing, pest control and other service businesses in Jonesboro and Northeast Arkansas with a few hundred online reviews and a rating that has slipped. If you read the one-star reviews one at a time and still cannot say what changed, this is that job.
The problem, in your words
Every bad review feels personal, and each one seems to be about something different. Was it the prices, the wait, one tech, or the office? You have a hunch, but a hunch is a hard thing to fix a business around, and nobody has the evening it would take to read two years of reviews and tally them up.
What happens, step by step
- Your reviews are collected into one list, with their dates and star ratings.
- Each review is tagged with the themes it mentions, such as no call back, a late arrival, a price surprise, or praise for a technician.
- The themes are counted against the star ratings, so you can see which complaint brings the low stars.
- The themes are tracked month by month, so you can see when a problem started.
- Any review that mentions a safety issue or a lawyer is kept out of the counts and sent straight to you to read.
What you approve
Nothing is posted and no review is answered. The result is a short report for you, and the reviews that need a careful response come to you first, by name.
What the example run showed
The video is a real run on fictional data: two years of made-up reviews, 500 in all, for a made-up heating and air company, tagged with plain keyword and phrase rules. “No call back” stood out: 76% of those reviews were one or two stars, and the share of reviews mentioning it jumped from 6% to 26% in November 2025, the month that change was planted in the data. About 92% of the theme tags were right when checked against hidden labels, and 11 reviews went straight to the owner. Those numbers come from one run on invented reviews.
Setting it up in your business
This is business automation that reads what your customers already wrote and tells you what to fix. It starts with a 20-minute call about where your reviews live and what you suspect. If it fits, a Workflow Review maps the job, and an Automation Pilot builds a report you can run again every month. I’m in Jonesboro and glad to meet in person.