The short answer
Not reliably. So does FAQ schema help AI Overviews? In the LVL3 Home Services Growth Index, sites with FAQ schema appeared in a larger share of their AI Overviews, but in our controlled test the gap held up only among mid-traffic sites and got unclear once we accounted for trade. Our verdict is mixed. I wouldn't make it your first fix.
Here's the thing. The markup is cheap to add, and the questions and answers it describes are the part a reader actually sees. Get those right first, then mark them up.
What the LVL3 Home Services Growth Index found
We crawled 464 home services websites on October 20, 2026, across HVAC, plumbing, electrical and roofing in 25 metros. For AI Overviews, we took the 397 companies with 50 or more AI Overview searches and measured how often each site appeared as a source in the Overviews its searches triggered, using Semrush data.
- 69 of the 397 companies (17%) had FAQ schema.
- Their median appearance rate was 5.7%. Companies without it: 4.1%.
- Across all 397, the median was 4.5%. Low-traffic sites sat at 2.7% and high-traffic sites at 5.7%.
Why a 1.6-point gap still reads as mixed
That raw gap looks like a win. It isn't a clean one. In our controlled test, the difference held up only among mid-traffic sites, and once we held trade constant it stopped being clear.
Trade matters here because the four trades don't adopt FAQ schema at the same rate. Across all 464 sites, about 13% of plumbing and electrical sites carried it, against 20% of HVAC sites and 25% of roofing sites. A raw comparison mixes all of that together.
So the honest reading is this: some sites with FAQ schema do better in AI Overviews, and our data can't tell you the schema is why.
How FAQ schema compares with the other AI shortcuts
We tested the shortcuts people sell as the way into AI search, with traffic and trade held constant. Of these three, FAQ schema was the only one with a mixed result rather than no lift.
- llms.txt file (26% of the 397 have one): no measurable lift in our sample.
- LocalBusiness schema (68% have it): no lift.
- FAQ schema (17% have it): mixed.
What does track with AI Overview visibility is overall search strength. The typical high-traffic site appeared in more than twice the share of its Overviews that the typical low-traffic site did (5.7% against 2.7%).
What Google says
Google cut back FAQ rich results in August 2023. Since then Google shows them mainly for well-known, authoritative government and health websites. For a contractor, the markup stopped earning the expandable questions in the regular results a long time ago.
For AI features, Google's own documentation says there are no additional requirements to appear in AI Overviews or AI Mode, and no special optimizations needed. The same SEO basics apply. If you use structured data, it should match what's visible on the page.
Neither statement says FAQ schema hurts. They say it isn't a key to a special door.
What I'd do instead
Three things need to happen, in this order.
- Write the questions people actually ask. Pull them from call recordings, CSR notes and Google's People Also Ask, then answer each one in plain language on the service page it belongs to.
- Mark up what's on the page. Add FAQPage schema to the questions and answers a visitor can read. Never mark up questions that aren't visible.
- Measure it. Track AI Overview appearances monthly in Semrush, so you know whether anything moved instead of guessing.
And hold off on llms.txt until the first two are done. It showed no measurable lift in our sample.
How we handle it
Our pages carry a direct answer up top, clean schema and quotable passages, so an engine can lift them without guessing. Each of the 45 location pages we rebuilt for True Food Kitchen between February and August 2026 has a nine-question FAQ written for that restaurant.
That case measures rankings and revenue, not AI Overviews. So I won't credit the FAQ with anything an Overview did. The point is simpler: the questions were worth answering on their own.
Sources
- The LVL3 Home Services Growth Index 2026: 464 operator websites crawled October 20, 2026. AI Overview figures cover the 397 companies with 50 or more AI Overview searches, from Semrush data. Medians are raw; the verdicts hold traffic and trade constant.
- Google Search Central blog, "Changes to HowTo and FAQ rich results" (developers.google.com/search/blog/2023/08/howto-faq-changes), August 8, 2023, accessed November 9, 2026.
- Google Search Central, "AI features and your website" (developers.google.com/search/docs/appearance/ai-features), accessed November 9, 2026.