The 5-step diagnostic
Step 1: pull yoursite.com/robots.txt. Confirm GPTBot, ClaudeBot, PerplexityBot, Google-Extended are not disallowed. Step 2: ask ChatGPT and Perplexity to "list 5 alternatives to [your competitor]" — if you're missing, you're either uncited or your description is too vague to match the query. Step 3: search for your brand name on G2, Reddit, ProductHunt, and 2-3 niche review sites in your category. Mentions on these sites disproportionately influence AI answers. Step 4: run your homepage through Google's Rich Results Test. Broken or missing schema = no structured signals for AI engines. Step 5: check if your product was launched after the model's training cutoff (most are 6–12 months behind). If yes, AI live-browsing is your primary path until the next training refresh.
The mention gap is the biggest predictor
Across the brands we've audited, the strongest predictor of AI visibility isn't on-site SEO — it's third-party mentions on authoritative review/list sites. AI training data oversamples G2, Capterra, ProductHunt, Wirecutter, niche industry blogs, and Reddit. A new competitor with one G2 listing and three Reddit threads will frequently out-cite a 5-year-old product with no third-party presence. This is the modern "link building" but the rules are different — quality and topical authority matter more than DR scores.
What to do this week
(1) Submit your product to 5 relevant directories and review sites. (2) Audit and fix robots.txt and schema. (3) Rewrite your hero section with specific outcomes. (4) Add a comparison page targeting your top 2 competitors. (5) Post a useful, non-promotional thread on the most relevant subreddit. We've seen this 5-step playbook move products from invisible to consistently cited within 4–8 weeks. Run our GEO readiness checker for a starting score.