Real numbers, including the boring ones.
The tint shop results are Tint Guy LLC, a client since September 2023, named with its permission. Every case below starts with a measured baseline of where the business actually stood. We publish the baseline and the after, we name a client only when they agree and never change the numbers, and we include the case where the client was already doing well and there was not much left to fix.
Tint shop results: Tint Guy LLC
Real form leads from Google Ads went from 17 to 42 a month and cost per lead from $31.77 to $13.32. Google reviews went from 483 in December 2023 to 1,008. ChatGPT, Perplexity and Claude name the shop for the best tint shop in town.
What the scores are, and what they are not.
Each number below is an AI Visibility Audit score: a measure of how findable and machine-readable a business is across Google, the map pack and the AI tools. It is a diagnostic, and you can run it on your own site with our AI Visibility Audit.
Being straight about the limits, because this is where most agency case studies quietly overreach. A score improvement is not a revenue claim. It measures whether the signals that make a business findable are in place. Revenue depends on your market, your pricing, your close rate and what happens when the phone rings, none of which a schema fix controls.
What we can show you is the work, the baseline, the result, and how long it took. What we will not do is attach a revenue figure we cannot trace, which is the standard practice in this category and the reason most case studies are worthless.
Three businesses, three different starting points.
| Business | Overall | Schema | Timeframe |
|---|---|---|---|
| Roofing contractor, North Idaho | 57 → 94 | 0 → 100 | 5 days |
| Window tinting, Tint Guy LLC | 66 → 90 | 75 → 100 | 1 week |
| Tree service, Middle Tennessee | 92 → 100 | already strong | days |
Anonymized by trade and region, consistent with how we report the 2026 study. Clients can be named on request; we default to not naming them.
A hundred-page site with zero structured data.
Roofing contractor, North Idaho. Years in business, strong local reputation, real Google reviews, a genuine website with real content. On paper a well-established local company, and it ranked reasonably well in traditional search.
The audit came back with a schema readiness score of zero. Not low. Zero. Across more than a hundred pages there was not a single JSON-LD block: no LocalBusiness schema telling AI platforms what the business does or where it works, no Service schema for any service offered, and no FAQPage schema despite FAQ sections sitting on multiple pages. Overall score 57, with six critical gaps.
The business was structurally invisible to AI platforms, and increasingly to the structured signals Google itself uses. Not because the work was bad, but because nobody had ever added the layer that machines read.
Overall score in five days. Schema readiness zero to 100, all six critical gaps closed.
Read the full case study900+ reviews and a zero percent AI citation rate.
Tint Guy LLC, window tinting, Murfreesboro, Tennessee. This one is the most instructive, because by every normal measure the business was doing everything right. Years in operation, more than 900 Google reviews, an active and updated Google Business Profile, and a site ranking for its target keywords.
The audit scored it 66. GBP alignment came back at zero: the profile was not connected to the business entity in a way AI platforms could verify. FAQPage schema was missing entirely despite FAQ content across multiple pages, and there was no Person schema connecting a named human to the content.
A business can have an excellent reputation with actual customers and still be unreadable to the systems now making recommendations. That is the gap the whole three-surface argument is about, and this is what it looks like in one business.
After one week. The owner’s message afterward: “It works! Whatever you did to our site last week has us now showing up in most of the major AI platforms.”
Read the full case studyWant to see where you actually stand?
Free 30-second check across Google, the map pack and the AI tools. No sales call.
The one where there was almost nothing to fix.
Tree service company, Middle Tennessee. We include this deliberately, because a page of dramatic turnarounds implies every business is broken and most are not.
This company audited at 92 out of 100 before we did anything. Technical health 100. Content 100. GBP alignment strong, authority signals solid. By almost any measure a well-optimized local site, and the honest thing to say is that they did not need much.
One critical gap remained: FAQPage schema was missing, so FAQ content across service and location pages was readable by humans but could not be extracted or cited directly by AI platforms. We closed it.
One gap, closed. If your audit comes back like this, we will tell you there is not much to do rather than invent a project.
Read the full case studyA recovery operation Google only knew by name.
Heavy haul towing and recovery, Middle Tennessee, four yards. This one is measured differently from the three above. Rather than an audit score, it tracks search position before and after a rebuild, using two twelve day windows that contain no launch date.
Their site ranked for the business name and almost nothing else. It had no page at all for “wrecker service”, a phrase people type at 2am when a trailer is across two lanes. We built the page the search data asked for, and twenty two more like it.
Average position across all searches, in twelve days. Page two to page one. The click-through rate fell over the same period, which we publish and explain rather than omit.
Read the full case studyPage five to page one, on a business with a season.
Watersports and boat-dock dealer, Middle Tennessee lakes. The site sold docks, lifts and jet ski ports at four figures apiece, and Google mostly knew it for “dock accessories” at position 95. Average position across all searches was 48, which is page five.
Because the business is seasonal, every figure compares the same calendar window a year apart rather than spring against summer, and the later window sits entirely after the engagement began.
Average position, same months one year apart. Searches in the top three went from 28 to 181. Three of the new first-place rankings produced zero clicks, which we publish and explain rather than omit.
Read the full case studyCost per lead cut 58%, and a counting problem worth more than the number.
Tint Guy LLC, automotive window tint, Murfreesboro, Tennessee. Six campaigns running, one paying $154 for a lead and another $45 to buy the shop’s own business name. We rebuilt the search side around a single campaign aimed at the service the owner actually wanted to sell.
Auditing the account then turned up something more useful than the result: eight separate trackers recording phone calls into one total, three of them recording the identical event. Cost per lead had been reading better than it really was.
Cost per verified form lead, month one to month three, on unchanged spend. We report the strict number, server-side form submissions counted once, and explain why the account’s own total was not it.
Read the full case studyAnd 64 businesses we have no relationship with.
Client results have an obvious selection problem: we choose which to publish. So we also measured businesses that are not our clients and had no reason to look good.
Sixty-four real local service businesses, eight trades, four metro areas, every one already ranking in its local map pack. We checked whether each appeared across Google search, the map pack and AI answers.
Appeared in all three places. One in five appeared in only one. Not a sample of struggling businesses; they all already ranked on the map.
Source: Tristar 2026 Local Visibility Study. Full method and raw data.About these results.
Why are most clients anonymized?
Some are white-label engagements where naming us would be inappropriate, and others simply prefer not to advertise who does their marketing, particularly in competitive local markets.
The numbers, trades and regions are real, and clients who are happy to be named can be. Tint Guy LLC is. We just do not make that the default.
Why not show revenue increases?
Because we cannot trace them honestly. A business’s revenue moves with pricing, seasonality, crew capacity, close rate and how fast they answer the phone. Attributing that to a schema fix would be a claim we cannot defend, even when the direction is real.
We show what we measured and did. If an agency shows you a revenue figure, ask exactly how it was isolated.
How long do results usually take?
Technical and schema work moves fastest, sometimes within days as these cases show, because it is a matter of adding signals that were absent. Profile and map pack changes typically take a few weeks. Competitive search rankings take months.
The fast wins above are real and they are the beginning, not the whole job.
What if my audit comes back already strong?
Then we tell you that, as in case three. Plenty of businesses are in better shape than they think and need one or two things closed rather than a program.
The free check gives you that answer whether or not you ever speak to us.
Run the same check these started with.
Free, 30 seconds, across Google, the map pack and the AI tools. See who gets the call instead of you, and a link to your result you can keep. No sales call.
Check My Shop Free →