AI Search Optimization for Small Business: A 2026 Tactical Guide
AI search is reshaping how customers find businesses. Here is the actual to-do list for a small business owner with limited time.
By Laird Doman · Updated May 2026 · Ogden, UT
What this guide covers
What AI search optimization is, briefly
AI search optimization (sometimes called AEO, GEO, or just "getting cited by AI") is the work of making sure AI systems can find your business, understand it accurately, and include you in the answers they generate for relevant questions. It sits alongside traditional SEO, not instead of it. The goal is not to replace your Google rankings; it is to show up in the growing share of queries that now get answered by a language model before a user ever clicks a link.
Three places this matters for a small business owner right now:
- AI Overviews in Google. These appear above the standard results for many informational and local queries. If your content is cited, you get a mention before the first blue link. If it isn't, you may be invisible to a significant slice of searchers.
- ChatGPT and Claude direct queries. A plumber in Roy, an accountant in Farmington, a dentist in Layton: people ask ChatGPT for recommendations by name, city, and service type. What it says about your business depends almost entirely on what is written about you on the web and in public data sources.
- Perplexity and other AI-search browsers. Perplexity is the fastest-growing research tool for small business buyers. It cites sources. Being one of those sources is increasingly worth the effort.
The underlying work is mostly about clarity and authority. You need your business described consistently across the web, your content structured so AI can parse it, and enough credible external mentions that models trust what they find. The GEO playbook covers the conceptual side. This guide covers what to actually do.
The this-week checklist
Five hours is enough to get a baseline and find your three biggest gaps. These tasks require no developer, no paid tools, and no prior SEO knowledge.
1. Set up your free monitoring tools (1.5 hours)
If you don't have these connected yet, start here. They cost nothing and tell you what Google and Bing already know about your site.
- Google Search Console. Go to search.google.com/search-console and add your site. Verify ownership via your domain registrar or the HTML tag method. Give it a few days to populate data. This tells you which queries bring traffic and which pages get impressions without clicks.
- Bing Webmaster Tools. Bing still drives a non-trivial share of AI Overview traffic (Microsoft's Copilot is built on it). Go to bing.com/webmasters and connect your site. You can import your GSC settings to save setup time.
- Ahrefs Webmaster Tools. Free version of Ahrefs for your own site. Shows crawl errors, broken internal links, and which pages have zero inbound links. Useful baseline before you start building content.
2. Audit your existing content for FAQ structure (1 hour)
AI systems prefer content that directly answers questions. Go through your top five most-visited pages and your homepage. For each page, ask: does this page answer a specific question, or does it just describe what you do? Pages that answer questions ("How long does a roof inspection take in Utah?") tend to get pulled into AI answers. Pages that list services without context tend not to.
Write down which of your pages answers a clear question and which don't. That gap list becomes your content calendar for the next 90 days. If you want a faster baseline, run the free AI Grader first.
3. Test 10 queries about your business in ChatGPT, Claude, and Perplexity (1.5 hours)
This is the most important 90 minutes you will spend on AI search. Open all three tools and run queries as if you were a potential customer who does not know your business name. Examples for a landscaping company in Layton:
- "Best landscaping company in Layton Utah"
- "Who does sprinkler installation in Davis County"
- "Landscaping company that does retaining walls near Kaysville"
- Your actual business name
Write down exactly what each tool says. Copy the responses into a document. Note whether you appear at all, whether the description is accurate, and what competitors or alternatives each tool mentions.
4. Document and prioritize (1 hour)
From your query tests, identify the top three inaccuracies about your business. Common ones: wrong service list, outdated address or phone number, missing specialty you actually offer, or a competitor being cited for work you do. Rank them by which one a real customer would notice most. Those three become your first content and citation corrections in weeks 1 to 4 of the playbook below.
The 90-day playbook
This is the main program. Each phase has a focused priority. Do not skip ahead. Foundations first, then authority, then measurement.
Weeks 1 to 4: Foundations
The goal this month is to make sure AI systems can read your business accurately. Most small businesses have at least a few citations that are wrong, outdated, or missing. Fix those first before building anything new.
Add structured data (schema markup) to your site
Schema markup is machine-readable metadata that tells AI crawlers what type of business you are, what you offer, where you are located, and how to contact you. At minimum, add LocalBusiness schema to your homepage and FAQPage schema to any page with questions and answers. If you're on WordPress, plugins like Yoast or Schema Pro handle this without code. If you're on a custom site, a developer can drop in a JSON-LD snippet in under an hour. The AI invisibility guide explains why this matters in more detail.
- LocalBusiness schema on homepage (name, address, phone, service area, hours)
- FAQPage schema on any page with Q&A format content
- Service schema for each major service offering
- Validate with Google's Rich Results Test before going live
Add an llms.txt file
This is a plain-text file at the root of your domain (yourdomain.com/llms.txt) that tells AI crawlers what your site is about and what they're allowed to index. It's a newer convention, not universally supported yet, but it costs nothing to add and several major crawlers already read it. Keep it brief: business name, description, list of main URLs, and any content you prefer AI not to use. More detail in the GEO playbook.
Fix inaccuracies in your Google Business Profile and key citations
AI models pull heavily from Google Business Profile, Yelp, Bing Places, industry directories, and any site with structured business data. Go through yours and verify:
- Business name, address, and phone match exactly across every listing (even minor differences confuse AI systems)
- Service categories are as specific as the platform allows
- Business description uses natural language that answers "what do you do and who do you serve"
- Google Business Profile Q&A section has at least three answered questions you wrote yourself
Then address the three inaccuracies you identified in the week-one checklist. If AI is citing the wrong phone number or a competitor for a service you offer, fixing the source data is the fastest correction available.
Write or rewrite your top three FAQ-style content pieces
Take the gap list from your content audit. Write one page per question, 400 to 600 words each. The format that AI systems pull from most consistently: a clear question in the H1 or H2, a direct one-to-two sentence answer right below it, then supporting detail. No fluff introduction, no "great question." Just the answer, then the context. Link these pages from your homepage and any relevant service pages. See the guide on testing AI against your business for examples of how to frame these answers.
Weeks 5 to 8: Authority building
By now your own site should be readable and accurate. The next constraint for most small businesses is that AI models have very little third-party information confirming you're real and credible. Three quality external mentions change that more than 30 low-quality directory listings.
Get listed in one legitimate industry directory
Not a generic SEO directory farm. A real one that is specific to your trade: the state contractor licensing board site, the local chamber of commerce member directory, a professional association (NAHB, AICPA, NAR), or an industry-specific platform like Houzz for contractors or Zocdoc for healthcare. These carry genuine authority because they have editorial standards. One good listing beats 20 weak ones.
Get a mention on a podcast or local publication
An appearance on a local business podcast, a quote in a regional news article, or a contributed piece to a trade publication all produce text on authoritative URLs that AI crawlers index. This doesn't require a PR budget. Most local podcast hosts are actively looking for small business owners to interview. A contributed article to a regional chamber newsletter often gets a web version. Even a quote in a local news story about your industry counts. Look for three to five of these over the 90 days, with at least one live by the end of week 8.
Get one expert-quote mention in a publication
Sign up for HARO (now Connectively) or Qwoted. Journalists regularly ask for expert quotes from local business owners. A plumber quoted in a home improvement article, a financial advisor cited in a personal finance piece, a dentist mentioned in a health column: these create authoritative citations that are exactly what AI models pull when building answers. Respond to five to ten requests this phase and aim to land at least one. See the guide on getting cited by ChatGPT and the Perplexity citation guide for more on this approach.
Weeks 9 to 12: Iteration
This phase is about measuring what changed and doubling down on what worked. Do not add more tactics until you have measured the ones already running.
Re-run the original 10 queries
Open the document from week one with your original ChatGPT, Claude, and Perplexity responses. Run the same queries again. Copy the new responses next to the old ones. You're looking for: do you appear where you didn't before, are the descriptions more accurate, did any competitors lose mentions. Even partial improvements are real signal that the work is landing.
Review GSC and Bing Webmaster data
By now you should have 60 to 90 days of data in Google Search Console. Look at which pages earned the most impressions and which earned the most clicks relative to impressions. Pages with high impressions but low clicks often have a title or meta description that doesn't match what the searcher wanted. Fix those. Pages with consistent traffic are your best candidates for deeper content expansion.
Double down on what moved
If the FAQ pages drove new traffic, write three more. If the podcast mention led to a citation, book another appearance. If your Google Business Profile update changed how ChatGPT describes you, refine the description further. The businesses that build compounding visibility in AI search do so by iterating on what actually works for their specific category and market, not by following a generic checklist forever.
What tools to use
You do not need to subscribe to everything. Start with the free tier and add paid tools only when you have a specific question they answer that you can't get elsewhere.
- Google Search Console Track which queries send traffic to your site and spot crawl or indexing issues.
- Bing Webmaster Tools Bing-specific data and Copilot visibility signals; also catches issues GSC misses.
- Ahrefs Webmaster Tools Crawl report and backlink baseline for your own domain. Free for site owners.
- ChatGPT / Claude / Perplexity Direct query testing. The most honest signal for how AI describes your business today.
- Doman AI Grader Free AI visibility scan. Shows where your business stands in under two minutes.
- AI Visibility Report ($99) Full audit of how ChatGPT, Claude, Gemini, and Perplexity describe your business. Delivered within one business day. Prioritized fix list included.
- Strategy Session ($999) 90 minutes with Laird, custom plan built around your business, two-week follow-up. In person across Northern Utah or virtual.
- Implementation / Retainer Doman AI also handles hands-on setup, custom builds, ongoing optimization, and coaching. Scope depends on what you need.
- Profound Tracks AI citation mentions across multiple models over time. Useful once you have a baseline to compare against.
- Otterly AI search monitoring with alerting for brand mentions. Good for businesses in competitive local categories.
- Peec AI Visibility tracking across Perplexity, ChatGPT, and Google AI Overviews. Earlier-stage tool but improving quickly.
Mistakes to skip
The AI search optimization space is new enough that bad advice circulates freely. Here are the common ones to avoid before spending any money.
Buying "AEO certification" courses
Answer Engine Optimization is not a credential. No course certification makes you better at it, and no vendor issuing certificates has any relationship with how ChatGPT or Perplexity actually rank sources. Save the money and spend those hours implementing the free steps in this guide instead.
Hiring agencies that promise guaranteed AI citations
No one controls which sources an AI model cites. Agencies that promise guaranteed mention placement are either misrepresenting what they're doing, gaming platforms in ways that could hurt you, or both. A legitimate AI search consultant shows you what they're doing and why, and sets realistic timelines.
Spamming forums and comment sections with brand mentions
The idea behind this tactic is that if AI models see your business name mentioned on Reddit, Quora, and other crawled forums, they'll include you in answers. In theory, authentic forum presence matters. In practice, posting fake reviews or planted mentions is detectable, violates platform terms, and produces low-trust signals that quality models are increasingly filtering out.
Spinning AI-generated bulk content
Publishing hundreds of thin AI-generated pages to capture every possible query variation is a short-term play that degrades your site's quality signals over time. Google's helpful-content systems and the citation quality filters in AI models both penalize low-value content at scale. Fewer, better pages consistently outperform content farms.
Ignoring on-site basics
A slow-loading site, broken schema, duplicate title tags, or a home page with no clear description of what you do: these undermine everything else. AI crawlers encounter the same technical barriers that Google does. Run a crawl report with Ahrefs Webmaster Tools and fix the flagged issues before adding new content.
When to DIY vs when to bring in help
Most of the week-one checklist and a fair amount of the first 30 days are genuinely doable on your own. Here is an honest read on when to stay solo and when to stop trying to figure it all out independently.
DIY if: you have five or more hours a week to spend on this, you're comfortable with a browser-based tutorial, your site is on a platform like WordPress or Squarespace where plugins handle schema, and you don't mind a slower start in exchange for learning as you go. The free grader, free monitoring tools, and this playbook give you a real starting path.
Get help if: your site is custom-built and schema requires a developer, you want to know where you actually stand before guessing at priorities, you have a competitive local market where getting this wrong costs real customers, or you simply don't want to spend weekends learning a discipline that changes quickly. Having someone audit your starting position is not a luxury if it means not spending six months optimizing for the wrong things.
Doman AI offers three structured ways in. The $99 Visibility Report tells you exactly how the major AI systems describe your business today and what the top fixes are. The $999 Strategy Session turns that into a prioritized custom plan with a two-week follow-up. Beyond those entry points, the engagement can go wherever makes sense: coached implementation where Laird walks you through the execution, full done-for-you setup, custom builds, an ongoing retainer, or just occasional check-ins when you have questions. Start with what fits your budget and your schedule.
Questions
How is AI search different from regular SEO?
Traditional SEO targets Google's ranked list of blue links. AI search targets the summary answers that ChatGPT, Perplexity, Claude, and Google's AI Overviews generate. The ranking factors overlap but are not identical. AI systems weigh authoritative, well-structured, factually consistent content more heavily than keyword density. Both disciplines matter and they are not either/or.
Should small businesses bother with AI search yet?
Yes, but with realistic expectations. AI search is not replacing Google traffic this quarter. It is growing, and the businesses that build good structured content and accurate citations now will have a head start when the shift accelerates. The good news: the foundations you build for AI search also strengthen traditional SEO.
How long until AI search efforts pay off?
Most owners see measurable changes in how AI models describe their business within 60 to 90 days of consistent effort. Full citation presence across ChatGPT, Perplexity, and AI Overviews typically takes three to six months. Quick fixes like correcting inaccurate citations can show up faster, sometimes within a few weeks.
Can I do this myself with limited tech skills?
Most of the week-one checklist requires no coding at all. Setting up Google Search Console, updating your Google Business Profile, and testing queries in ChatGPT are all browser tasks. The schema markup step is the most technical and can be handled with a WordPress plugin or a simple JSON-LD snippet your web person drops in.
What is the cheapest way to start?
Free: set up Google Search Console and Bing Webmaster Tools, verify your Google Business Profile information is accurate, and run the Doman AI Grader to see where you stand. That takes about two hours and costs nothing. If you want a structured baseline report, the $99 AI Visibility Report is the next step.
Can Doman AI help me?
Yes. The $99 Visibility Report shows exactly how the major AI systems describe your business today and what to fix. The $999 Strategy Session turns that into a prioritized 90-day plan built around your specific business. Beyond that, Doman AI can handle implementation, build custom tools, manage ongoing optimization, or just coach you through the process at whatever pace works for your schedule.
Related playbooks
Also relevant: Why AI invisibility costs businesses and Try AI on your own business.
See where you stand today
The free Grader takes two minutes. The $99 Visibility Report gives you a full picture with a prioritized fix list. If you want a custom plan, the Strategy Session is 90 minutes with Laird and walks out with a real 90-day roadmap for your specific business.
Or try the free AI Grader first →
Questions? laird@domangrowth.com · Based in Ogden, UT