Get named in 30 days: an AI search checklist for ecommerce brands

You don't need a team or a budget to get started with AI search. You need a list of questions, a spreadsheet, and a few hours a week for a month.

This checklist walks you through it, one week at a time. Each week builds on the one before, so do them in order. If you want the background on why each step matters, read the full guide first.

Before you start: set up your tracker

Make a spreadsheet with these columns. You'll fill it in during week 1 and come back to it on day 30.

• Question: Best running shoes for flat feet under $150? · ChatGPT: Named · Claude: Absent · Gemini: Absent · Perplexity: Named · Who was named instead: Brand A, Brand B · Sources cited: Review site, Reddit thread · Page that should answer it: /collections/stability

For each engine, mark Named or Absent. Keep it simple. You're looking for patterns, not precision.

Week 1: Listen

Goal: know exactly what AI says about your category today.

Collect your questions

• Pull 10 questions from support tickets and chat logs
• Pull 10 from product reviews (look for use cases: "I bought these for...")
• Pull 5 to 10 from your site search report. How-to: How to find what shoppers search for on your store
• Ask your customer service or sales team for the questions they hear before people buy
• Rewrite each one the way a shopper would ask it out loud ("best running shoes for flat feet under $150," not "flat feet shoes")
• Cut the list to your 20 to 30 most important questions

Run them across all four engines

• Ask every question in ChatGPT, Claude, Gemini and Perplexity
• Use a fresh or logged-out session so your history doesn't color the answer
• Record Named or Absent for each engine
• Record which brands were named instead of you
• Record which sources were cited (your site, retailers, review sites, Reddit, publications)

Sort the results

• Pile 1: questions where you're named. Protect these.
• Pile 2: questions where a competitor is named and you're not. This is your to-do list.
• Pile 3: questions where the answer is vague or wrong. Note what's wrong.

Short on time? The free Selix check runs your store across all four engines for you. Free. No account. 45 seconds.

Week 2: Fix the foundations

Goal: make sure AI can read your site and trust your product data.

Check that you aren't blocking AI

• Open yourstore.com/robots.txt. How-to: How to check your robots.txt for AI crawlers
• Make sure these search bots are allowed: OAI-SearchBot (ChatGPT search), Claude-SearchBot and Claude-User (Claude), PerplexityBot (Perplexity)
• Decide separately whether to allow the training bots, GPTBot and ClaudeBot. Blocking them doesn't remove you from search answers.
• Check your CDN and security settings for bot-blocking rules that could catch AI crawlers. How-to: How to check if your security settings block AI crawlers

Make your key pages readable

• Pick your 10 most important product pages
• On each one, view the page source and search for a key detail (a size chart line, a spec, a fit note). How-to: How to check if AI can read your product pages
• If it's missing from the source, it's loaded by JavaScript. Move the essential facts into the page itself.
• Add a short written summary of what reviewers say, in plain text on the page

Clean up your product data

• Rewrite vague titles so they say what the product is ("Women's Stability Running Shoe, Wide Fit")
• Fill in every product attribute your platform offers: sizes, widths, materials, dimensions, compatibility, care
• Run your top pages through a schema validator and fix missing price, availability or rating fields. How-to: How to check your product structured data
• Compare your site, Amazon and retailer listings, and fix any mismatched prices or specs

Week 3: Write the answers you're losing

Goal: publish a clear answer to your five most important Pile 2 questions.

Pick your five

• From Pile 2, choose the five questions closest to a purchase
• For each one, decide which page should answer it (product page, collection page, FAQ or guide)

Write each answer

• Put the answer near the top of the page, in one or two plain sentences
• Say who the product is for and what it's good at ("a firm arch and a wide toe box, which makes it a strong choice for runners with flat feet")
• Say what it's not good for. Engines favor sources that sound like advice rather than ads.
• Turn at least one spec table into sentences that explain what each spec means for the shopper

Add an FAQ block

• Add an FAQ to your top collection page using real questions from your list
• Answer each in two or three sentences
• Read everything back once and cut anything that sounds like filler

Week 4: Earn mentions and measure

Goal: start building outside proof and get your baseline numbers.

Earn mentions outside your site

• Email recent customers asking for a review that mentions what they bought it for
• Look at the sources column in your tracker and find the review site, list or thread that shows up most
• Pitch the writer of one roundup that AI already cites in your category
• Answer one real question in a community where your shoppers hang out, as yourself, with no fake accounts

Set up measurement

• In your analytics, create a channel for AI referrals: chatgpt.com, perplexity.ai, gemini.google.com, claude.ai. How-to: How to track AI traffic in Google Analytics and Shopify
• Note which pages those visitors land on

Day 30: Re-run and compare

• Run your full question list across all four engines again
• Count your answer share: questions where you're named on at least one engine, divided by total questions. Count each question once.
• Compare against week 1. Which questions moved? Which didn't?
• Check whether AI referral traffic landed on the pages you updated

Don't expect every answer to flip in 30 days. Engines that search the web live can pick up changes once they recrawl, but other answers move more slowly. What you're looking for is a trend.

After day 30: keep going

• Re-run your question list every month
• Write answers for the next five Pile 2 questions
• Add new questions as they show up in support tickets and reviews
• Keep one outreach effort going each month (a review ask, a pitch, a community answer)

AI search isn't a one-time project, just as SEO never was. The brands that keep checking are the ones that stay on the list.

AI names a short list. Be on it.

Want to skip the spreadsheet? Check my store. Free. No account. 45 seconds.