For a local service business, AI visibility ROI comes down to one question. How much extra revenue did AI mentions bring in, compared to what you spent getting mentioned? The two things that make it hard to answer are getting a credible number for how many inquiries came from an AI recommendation, and knowing what a normal AI visibility score even looks like. Both are covered below, with real numbers, not guesses.
Why the Standard AI Visibility ROI Formula Doesn't Work for Local Service Businesses
There’s no single standard ROI formula for AI visibility, and the ones that do get published skew toward ecommerce or B2B thinking, neither of which fits a local service business. The ecommerce version is average order value times conversion rate. The B2B and SaaS version chains together lead rate, qualification rate, close rate, and contract value, built around a multi-stage funnel a CRM tracks automatically.
A local service business runs on far fewer, higher-stakes decisions, most of them closed in a single phone call or an in-home estimate. There’s no cart to abandon and come back to, no multi-stage funnel, no CRM logging every touch, just a call, a quote, and a yes or no, often in the same conversation. Force that pattern into an ecommerce or SaaS ROI model and the formula looks rigorous on paper, then falls apart the moment you plug in real numbers from a service call log, which is exactly why this vertical needs a formula built around how it gets paid, not one borrowed from a different business model entirely.
The Local Service ROI Formula, With Real Numbers by Industry
Here’s the formula in full.
Plug in numbers that match your industry, not a blended average pulled from nowhere in particular, and the number means something.
Say an HVAC company traces 40 qualified inquiries to AI mentions in a quarter, closes 25% of them, and the average repair ticket is $300. That’s 10 closed jobs, or $3,000 in AI-attributed revenue. Against $500 in AI visibility spend that quarter, the return is 500%. Change any single input and the number swings fast, which is why the industry-specific ranges below matter more than one blended average.
One honest caveat. Job value here means the ticket price, not profit after parts, labor, and overhead, so this is a revenue-based ROI. Most marketing ROI models work this way because businesses rarely hand over margin data, but it means the real, profit-based return on any job is lower than what this formula shows. Swap in your actual margin per job instead of the ticket price for a stricter number. The formula works the same either way.
HVAC numbers come from HomeGuide's repair and installation cost data, legal comes from Foundry CRO's benchmarks, dental comes from Crimson Media Group and Dandy, and the HVAC close rate comes from Invoca's 2025 home services benchmark report.
How to Attribute an Inquiry to AI Visibility Without an Enterprise Tech Stack
This is the part almost every other guide on this topic either skips entirely or assumes you have GA4 event tracking and a CRM sophisticated enough to tag AI referrals automatically. Most local service businesses have neither. Here’s the version that works without them, in order of how fast you can put each one in place.
1. Fix the intake question first
"How did you hear about us" is nearly worthless if the answer options are "internet," "search," or "referral." Replace it with something that separates AI platforms out by name. On a phone script, that's as simple as asking "Did you find us through a regular Google search, a friend, or an AI tool like ChatGPT or Gemini?" On a web form, swap the generic dropdown for one that lists ChatGPT, Gemini, Perplexity, and Google AI Mode as their own options alongside Google, Google Maps, and referral. This costs nothing, takes about ten minutes to change, and is the single highest-leverage fix most businesses can make this week. It's also the only method here that gives a direct, self-reported answer instead of an inferred one.
2. Track branded search as a secondary signal
AI answers rarely include a clickable link, so someone who just got a business recommended by ChatGPT or Gemini usually does the next logical thing. They Google the business name directly instead of clicking through. The result is a measurable bump in branded search volume, visible in Google Search Console's Performance report filtered to queries containing the business name, or in Google Business Profile's own Insights panel under direct/branded searches. It's an inferred signal, not a confirmed one, but a branded-search bump that consistently follows an uptick in AI mentions is a real correlation worth logging month over month.
3. Correlate Visibility Score against inquiry volume over time
The most complete picture comes from lining up two trend lines side by side. One is a business's Visibility Score, the metric Local Dominator's AI Tracker uses to measure how often and how prominently AI platforms mention it. The other is inquiry volume over that same window. Say a business's score climbs from 4% to 11% over eight weeks while phone inquiries citing "an AI assistant" on the new intake question rise from 2 to 9 over the same stretch. That's not proof any single call came from any single AI mention. It's a credible before-and-after read that the two are moving together, which is the level of confidence most local businesses need to justify the spend. Visibility Score is one of several AI search visibility metrics worth tracking alongside inquiry volume before drawing a conclusion from either number alone.
What a Realistic AI Visibility Score Actually Looks Like
Before any of this math means anything, it helps to know what a normal starting point looks like, because most business owners expect either near-zero or near-perfect and both expectations are wrong.
Local Dominator’s own state of local AI visibility report audited 12,535 domains across 1,071 campaigns, running 398,507 individual scans. The average visibility score across that entire dataset was 8.24%, with an average share of voice of 19.15% pooled across all results and 18.74% when every business is weighted equally. In plain terms, the typical local business is mentioned by fewer than one in ten AI engine queries relevant to it.
This number resets expectations before the ROI conversation even starts. A business sitting at 3% or 5% visibility a few months into AI visibility work isn’t failing. It’s tracking close to where most of the market already sits, which is a very different story than the one a low number tells on its own.
What Our Own AI Tracker Data Shows, By Service Category
The report above is a broad snapshot across every industry it touched. Pulling from Local Dominator’s own live AI Tracker campaigns and breaking the same metric down by niche shows a similar pattern, but with real spread between categories.
Visibility Score by Category
Current average Visibility Scores by category, drawn from active client campaigns and stated by niche only:
These are current snapshots from an active book of business, not a random sample of an entire industry, and should be read as directional rather than as a fixed industry law. What they do line up with is the state of local AI visibility report's own overall average. Most local businesses start low, across the board, which corroborates the broader dataset rather than contradicting it.
Inside the 120-Scan Mini-Study
To go a layer deeper, we ran a small original study for this piece, pulling 120 individual scan results from active HVAC-adjacent client campaigns. Four things stood out.
Mention rate. Only 1.7% of those scans, 2 out of 120, surfaced a tracked business by name. The other 98.3% were an AI platform answering an HVAC-related question with no apparent awareness the business existed at all.
By engine. Gemini surfaced a business in 2 of its 30 checks in this sample. ChatGPT, Perplexity, Google AI Mode, Claude, and Grok didn't surface one in any of theirs. Those are small per-engine counts, so read them as a snapshot of this sample, not a permanent ranking of which engines favor which industries.
Sentiment. Across all 120 scans, the HVAC-related answers themselves broke down as 76 neutral, 44 positive, and zero negative. Of the four findings, this is the most useful one. AI isn't biased against this industry, it's under-informed about who's in it, which reframes the whole ROI conversation from a reputation problem into a visibility gap, a far easier pitch to a client or a boss.
Most-cited sources. The domains AI cited most often across these scans were Angi, Google's own listings and Maps data, Yelp, HomeAdvisor, and BBB, plus a handful of niche HVAC-specific sites. It's directly actionable too. A business that wants a real shot at being cited should focus on getting listed and reviewed on exactly those platforms.
AI Visibility ROI vs. Traditional SEO ROI: How the Timelines Compare
Anyone working through this formula is also, almost certainly, already tracking traditional Google ranking ROI, and the two timelines don’t move at the same speed or answer the same question. Our local rank tracker guide covers how to prove ROI on the Google Maps and organic-ranking side using geo-grid data, while this one covers the AI-search side specifically. One shows where a business ranks. The other shows whether AI platforms even know it exists. Put together, that’s the full client-facing ROI story.
Turning This Into a Report a Client, or Your Boss, Actually Trusts
Having the formula and the attribution method is one thing. Getting a client or a boss to actually believe the number is another, and that’s where packaging matters as much as the math itself.
The formula, the attribution method, and the benchmark data above are only convincing when they show up somewhere a client can see for themselves, not buried in a spreadsheet only the agency ever opens. Local Dominator’s White Label Reporting feature turns the Visibility Score, the scan history, and the attribution signals into a branded report hosted on the agency’s own domain, so a client sees a coherent, professional deliverable instead of a screenshot pasted into an email. White Label Reporting is available on Local Dominator’s Pro plan.
Ready to see where your own AI visibility stands? See if ChatGPT, Gemini, and Perplexity are recommending your business.
Frequently Asked Questions
How long does it take to see ROI from AI visibility optimization?
There’s no fixed timeline, and anyone promising one is guessing. Most businesses start well below 10% visibility, so expect a gradual climb measured in months, not days, and track the trend line rather than any single scan.
What's a good AI visibility score for a local business?
Based on Local Dominator’s own dataset across more than 12,000 domains, the average is 8.24%. A business in the mid-teens or higher is already ahead of most of the market. A business near zero isn’t unusual, it’s the median starting point.
Can I track which leads came from ChatGPT specifically?
Not with session-level precision unless you’re running enterprise-grade attribution tooling most local service businesses don’t have. What works instead is naming AI platforms explicitly in your intake question, watching branded search lift, and correlating Visibility Score movement against inquiry volume over time.
Is the ROI formula different for ecommerce businesses than for local service businesses?
Yes, meaningfully. Ecommerce ROI models run on average order value and conversion rate through a cart. Local service businesses run on a smaller number of higher-value jobs closed mostly by phone or in-person estimate, which is why the formula in this guide uses job value and close rate instead.
How do I report AI visibility ROI to a client who doesn't understand SEO data?
Lead with the plain-language version of the formula and the dollar number it produces, not the underlying metrics. A branded, client-facing report that shows the Visibility Score trend alongside attributed inquiries does more to build trust than a raw data export ever will.
