Concrete has two customers who look nothing alike. One is the general contractor or builder sourcing a flatwork or foundation sub for a project, a repeat, professional buyer. The other is a homeowner who just noticed a crack running up the foundation and is quietly panicking. Both now start with AI. 45% of US consumers used AI tools like ChatGPT for local business recommendations in the past year (BrightLocal, 2026), and builders lean on it just as readily to shortlist subs. The mistake is writing for only one of them. The wins come from serving each on its own terms.
Who actually asks AI about concrete work?
On the consumer side, three jobs drive most of it: driveways ("cost to replace a concrete driveway"), patios and flatwork ("stamped concrete patio cost"), and the anxious one, foundation ("is this foundation crack serious", "foundation repair cost"). On the professional side, a GC asks "reliable concrete subs for a commercial slab in Sacramento" or "foundation contractors for a custom home build". These are different questions with different stakes. The homeowner is scared and price-curious; the builder is filtering for reliability and scope fit. You can be named for both, but not with the same page.
Why is 'is my foundation crack serious' the question to own?
Because it's high-anxiety, high-ticket, and almost no concrete contractor has answered it well. A homeowner staring at a crack doesn't know if it's cosmetic settling or a structural emergency, and that fear sends them straight to AI for a neutral read. Foundation repair is one of the most expensive jobs a homeowner ever faces, so the stakes on getting named as the trustworthy explainer are huge. The contractor who publishes a clear, honest diagnostic, when a crack is cosmetic, when it's structural, when to worry, becomes the source AI quotes and the name it hands over once the answer turns serious.
How do you get on a builder's subcontractor shortlist through AI?
By making your scope and reliability legible in text a model can quote. A builder asking AI for concrete subs is filtering on fit: do you do commercial slabs, structural foundations, decorative flatwork, tilt-up, and at what scale. A site that just says "concrete services" gives the engine nothing to match against a specific project. Name the work you actually do, the project sizes you handle, the cities you cover, and back it with real, dated jobs. The B2B side of concrete is underserved on purpose, because most subs assume they get work by word of mouth. The ones who write it down get found.
Who does AI name for concrete work?
The big platforms and the contractors who look complete on them. In Local Dominator's analysis of 267,280 AI citations drawn from its own clients' local-marketing campaigns, Yelp, Google, Reddit, Facebook, and Angi led the citation counts, with BBB and HomeAdvisor further down. It's a vendor's convenience sample, so lean on the ordering rather than the raw counts. For a concrete contractor the read is the same as for any local trade: complete, review-rich profiles on the platforms the engines trust, plus your own pages that state scope and price in plain words, is what gets you named.
What should a concrete contractor publish to get cited?
- A foundation-crack diagnostic guide for your area: cosmetic vs structural, soil factors, and when to worry.
- Priced pages per consumer job (driveway replacement, stamped patio, foundation repair) with honest ranges and what moves the price.
- A subcontractor capabilities page naming the work you do for builders (commercial slabs, structural foundations, tilt-up, decorative) and your project scale.
- A service-area page listing the cities and counties you pour in, so the engine can match you to a location.
- Dated project pages with the city, the job, the scope, and photos; specific is citable, 'quality concrete work' is not.
How do you measure it?
With a fixed set of your market's concrete questions, both consumer and builder-facing, asked repeatedly on every engine, tracked as the share of answers that name you. Each engine keeps its own sources, so being named on one says little about the rest: Profound found ChatGPT and Perplexity overlap on only 11% of the domains they cite. And the pool is narrow everywhere: SOCi's 2026 Local Visibility Index found ChatGPT recommended just 1.2% of local brand locations, versus 11% for Gemini and 7.4% for Perplexity. SOCi measured multi-location brands, not contractors, but the shape holds. Baseline, publish, re-measure.
