AEO Case Study: What a Utah Roofing Company Got From Answer Engine Optimization

What Were the AEO Case Study Results for This Roofing Company?

Over four months, AI engines went from never naming Olympus Roofing to naming it in close to a third of answers to the questions Salt Lake homeowners ask when choosing a roofer. Its own website also became one of the sites AI engines cite most. Each result below has its month, because AI answers change and a number without a date can't be checked.

Source: Wudel AI tracking and Olympus Roofing monthly reports, June to October 2026; summary at wudelconsulting.com/work/olympus-roofing.

What Did the AEO Program Actually Involve?

The program had three parts: a fixed list of customer questions tracked every day, answer-first articles published three times a week aimed at those questions, and technical work so AI engines could read and trust the site. Nothing in it was paid placement; it was all on Olympus's own website.

Tracked questions

We started with 30 questions a Salt Lake homeowner asks when choosing a roofer, worded the way people type them into ChatGPT: "best shingle roof replacement Utah", "best roof replacement companies Sandy", and so on. Those questions were asked of multiple AI engines every day. That gave a baseline and a way to see which competitors AI named instead.

Answer-first articles, three a week

Each article targeted one tracked question, put the direct answer in the first paragraph, and was published on a Monday, Wednesday or Friday. For "who should I hire" questions we wrote decision guides that compare several real local roofers with facts a homeowner can check, including Olympus's competitors. In October, those decision guides were the Olympus pages AI engines cited most.

Schema and technical work

Each article carries structured data (Article, FAQPage and, for comparisons, ItemList), using the schema.org vocabulary that search engines read. We added named author pages, an llms.txt file and internal links between related articles. Google's own guidance on AI features in Search says pages must be indexed and eligible for Search to appear in AI Overviews, so the basics came first.

What Didn't Move, and What Are the Limits?

Not everything moved, and some results can't be generalized. Olympus is one client, in one city, in one trade. AI answers vary from day to day, so a single day's number can mislead. And after four months, AI engines still named other roofers more often for many questions.

1. It is not #1. In October Olympus had 20.0% share of voice, third of nine tracked roofers. Two competitors were named more often.
2. Seven in ten answers still didn't name it. A 30% visibility rate in October means most AI answers to its questions named someone else.
3. Some questions stayed with competitors. In September, 3 of the 30 tracked questions never named Olympus. Its October report lists specific questions where AI still names other roofers, such as certified-installer and suburb-specific searches, as next month's work.
4. The question set changed. The early set mixed residential and commercial questions. From November 2026 tracking moved to a residential-only set, so November's numbers will not be compared directly with October's.
5. Tracking has gaps. Daily tracking had a gap in late September 2026; September figures use all days that were recorded.
6. Correlation, not proof. We can show that visibility rose after the articles went live, and which pages were cited. We can't isolate how much came from each article versus other factors, such as reviews and listings.

How Should You Judge Any AEO Case Study?

Judge an AEO case study by whether you could check it yourself: a named client, dated numbers, a stated baseline, the engines and questions measured, and what didn't work. If a case study gives a percentage increase with no starting number and no time frame, you can't tell whether it means anything.

Source: Wudel Consulting's reporting standard, applied to its own Olympus Roofing case study above.

The same rules apply to us. If you want to see how answer-first content gets cited in the first place, our guide on how to get cited by ChatGPT covers the mechanics, and what AEO is explains the terms.

Would AEO Work the Same Way for Another Roofing Company?

Probably not in the same numbers, but the method carries over. A roofer in a different city faces different competitors, different review sources and different questions. What transfers is the process: measure what AI says now, answer the exact questions customers ask, publish on a steady schedule, and report against the baseline every month.

Market matters. A city where several roofers already publish detailed, answer-first content will be harder to move than one where nobody does. That is why we start every program with a baseline, not a forecast. If you're comparing providers, our guide to the best AEO agencies in Utah lays out what to check.

See Where Your Roofing Company Stands

Want the same baseline for your business? Our free AI visibility check asks the AI engines your customers' questions and shows who they name today.

Frequently asked questions

• What is AEO for a roofing company?

  AEO (answer engine optimization) for a roofer is the work of getting the company named and cited when homeowners ask AI tools like ChatGPT, Gemini or Perplexity who to hire, what a roof costs, or whether to repair or replace. It means answering those exact questions on the roofer's own website, with structured data, and measuring whether AI answers start naming the company.

• How long did it take Olympus Roofing to see AEO results?

  The first measurable change came in the first month. Olympus Roofing went from 0% AI visibility in June 2026 to 29% in July 2026, when its first articles went live. Larger gains in site citations came later: 430 answers cited its site in September and 2,626 in October 2026. One client's timeline is not a promise for another.

• Which AI engines were tracked in this case study?

  Five AI engines were tracked daily in October 2026: ChatGPT, Gemini, Perplexity, Google AI Overviews and Microsoft Copilot. From November 2026, Olympus's tracking covers ChatGPT, Gemini, Perplexity, Google AI Overviews and Google AI Mode for a new set of residential questions, so later results will be reported against a new baseline.

• What is AI visibility?

  AI visibility is the share of AI answers, across the tracked questions and engines, that name the business. If 1,000 answers are collected in a month and 300 of them name the roofer, visibility is 30%. It differs from citations, which count answers that link to the business's own website as a source.

• Did the articles that got cited share a format?

  Yes. The Olympus pages AI engines cited most in October 2026 were decision guides, such as best-roofer and best-commercial-roofer comparisons, that answer a "who should I hire" question directly and compare several real local companies with checkable facts.

• Can a roofing company do AEO without an agency?

  Yes, if it has the time. The core work is choosing the questions customers ask, publishing a direct answer to each on the website, adding structured data, and checking the AI answers regularly. The hard parts are consistency, three articles a week for months, and measurement, since answers change daily and one manual check proves little.

• Why does a vendor's own case study need a disclosure?

  Because the vendor benefits if you believe it. A disclosure tells you who ran the work and who wrote the account, so you can weigh it. Look for a named client, dated numbers, a baseline and stated limits. If those are missing, treat the case study as marketing rather than evidence, whoever wrote it.