Every Built Local client has the same chart pinned in their admin dashboard: weekly Google review count for the last 90 days. Not total reviews. Not average rating. Just the velocity - how many came in this week, last week, the week before.
We track this because it's the single best leading indicator of map-pack movement. If velocity goes up, position follows. If velocity drops, position will drop too, within 4-6 weeks.
The pattern we see in every trade
Across roughly 40 clients we've watched the same shape:
- Month 1: Excitement. New site, new GBP, founder asking every customer for a review. Velocity peaks at 4-6 per week.
- Months 2-3: Velocity stabilises around 2-3 per week. Map-pack position climbs steadily.
- Month 4: The dip. Founder forgets to ask. Velocity drops to 0-1 per week. We have to intervene.
- Months 5-6: With intervention, velocity recovers to 1-2 per week, steady state. Without, it stays at 0-1 and rankings slide.
The month-4 dip is so consistent we now build a reminder into the dashboard: at week 13, the client gets an automated nudge ("review velocity has dropped - here are 8 customers you haven't asked yet").
The script that actually gets replies
The single biggest leverage point is the ask. We've tested seven variants of "please leave a review" with clients. The version with the highest reply rate (37% in our most-recent measurement, vs 8% for the agency-template version) is sent by SMS within two hours of job completion. The text:
"Hi $name, $founder here from $business - thanks again for today. Could you do me a quick favour? A 30-second Google review really helps me reach more local customers. Link here: $link. If anything wasn't right, hit reply instead and I'll fix it. Thanks!"
The "if anything wasn't right, hit reply" line is the underrated bit. It catches the unhappy customer before they leave a 1-star review, and it signals that you're not just review-mining. Reply rate goes up, and the rare bad-review interception is worth the entire system on its own.
The link
Most trades send customers a generic Google Maps URL. The reply rate is poor because the customer has to find your business, then find the "Write a review" button. We use the direct review URL Google provides in the GBP dashboard (it deep-links straight to the review composer) and shorten it with a custom path on the client's own domain (e.g. $client-domain.co.uk/r). The custom-domain short link gets ~40% more clicks than a bit.ly equivalent because customers trust the brand domain.
The response template
We script three reply templates for every client - one for 5-star, one for 3-4-star, one for 1-2-star - and the founder personalises them in 30 seconds each. Every reply mentions the suburb or street, which compounds local relevance for the UK local web design pages targeting those areas. A response saying "Thanks Maria, glad we could sort the Gosforth boiler!" reinforces the Gosforth signal in the same way a citation does, except more usefully.
What this looks like in the dashboard
The chart in the dashboard isn't just a count - it's overlaid with map-pack position for the client's top three keywords. The correlation is visible by eye. Velocity up by 1/week → position up by 1-2 within a month. Velocity flat → position flat. Velocity down → position down with a lag.
It's the most-looked-at chart in the entire dashboard, by clients themselves. They get the feedback loop. They ask for more reviews. The system feeds itself.
Why this is core to local web design for tradespeople
A local web design build that doesn't ship with a working review-request system is doing half the job. The site converts traffic; reviews bring the traffic. Without the review machine, you're paying for a site that ranks for nothing.
Every Built Local build ships with the SMS review request, the custom short link, the response templates, and the dashboard chart. See how it all comes together or read the GBP optimisation guide for the full picture.