An AI answer mentions your brand. That looks like progress, but it does not tell you what to do next. A competitor may appear more often, the answer may describe an old version of your product, or important pages may never reach the bots collecting information.
In this webinar, Constance Tan, Product Marketer at Ahrefs, explained how to separate those problems and choose the right response. Her starting point: look for recurring patterns across the customer journey, not a reassuring mention in one prompt.
Track Questions That Reflect How Customers Choose
Generating hundreds of prompts does not guarantee useful coverage. Tan organized monitoring around three kinds of questions:
- Problems customers need to solve. These capture discovery before someone knows which brand to consider.
- Positioning and comparisons. These reveal which brands AI recommends for particular audiences, use cases, and categories.
- Facts about your business. These test whether answers accurately describe pricing, capabilities, availability, and product use.
Search queries, support conversations, sales questions, and relevant community discussions can supply the language for those prompts. Tan cautioned that the most useful forums differ by market; Reddit and Quora are not the answer everywhere.
Constance Tan described the goal this way:
“The idea is that you want a representative view from every angle of the customer journey.”
Monitor that set over time. Repeated concrete to investigate
Follow Competitor Mentions Back to Their Sources
If competitors appear more often, share of voice identifies the gap. Reading the answers helps explain it. Does AI repeatedly favor another brand for small businesses or ecommerce? Which pages supply those distinctions, and does your content explain the same use cases?
Tan recommended examining both thetter alongside articles. For outreach, prioritize pages cited repeatedly and authors you can realistically reach. Domain strength and organic traffic add context, but an influential competitor-owned page may offer little opportunity for a correction
The recording’s competitive-ity comparison and narrows it down to the specific
Fix Inaccurate Information Where You Have Control
Before commissioning another article, inspect your own pricing pages, product explanations, profiles, and older posts. Conflicting information can leave AI answers describing features or plans that have changed.
Third-party corrections require more patience. Tan shared an Ahrefs outreach example: the team contacted 26 authors about inaccurate information, 10 replied, and four updated their content. One issue concerned older descriptions of which plans included API access.
Those are outreach outcomes, not proof of a corresponding lift in citations or revenue. They illustrate why selecting reachable, frequently cited
Updating someone else’s page is not always practical. Constance Tan explained the alternative:
“Sometimes outreach is not always the answer. Sometimes it’s better to create the newn a better way, a more comprehensive way, or with more up-to-date information.”
In the Q&A, Tan expanded on that choice: a warm relationship can make a correction worthwhile, while an important topic with weak coverage may justify an original guide or collaboration. If the same error appears across several
Earn Useful Mentions, and Check Bot Access
Tan’s advice for community participation was not to insert a product pitch into every thread. Answer technical questions, correct factual mistakes, or offer useful guidance. Recurring complaints can also reveal product or onboarding problems worth taking back to the teams that can fix them.
Missing citations can have a different cause entirely: bots may be unable to retrieve the content. Tan recommended checking firewall restrictions, broken URLs, timeouts, and pages that rely on JavaScript to display important information.
Investigate those failures before treating every visibility gap as a content problem. A useful page cannot serve as a retrieved
Turn the Findings Into Your Next Round of Work
Visibility reporting also needs business context. Asked about revenue, Tan discussed Ahrefs’ self-reported discovery data, including customers who mentioned ChatGPT, rather than claiming a revenue-per-citation formula. Her recommendation was to consider impressions and share of voice alongside conversions, sales, and customer attribution information.
Watch the full session for thet the approach into practice, start with one customer segment and use what you find to choose a specific action:
- Build a balanced prompt set. Cover customer problems, comparisons, and factual questions using language from search and customer conversations.
- Investigate repeated claims. Identify the sources behind recurring recommendations or errors instead of reacting to every isolated answer.
- Correct owned information first. Update outdated product and pricing explanations, then prioritize third-party corrections you can realistically secure.
- Match the fix to the problem. Use outreach for reachable sources, useful new content for coverage gaps, and technical checks for retrieval failures.
- Review visibility with business results. Track patterns over time alongside conversions and customer feedback, without treating a citation as a sale.
Join Us For Our Next Webinar!
A New Place To Look: Where Your Next AI Citations & Clicks Come From
Join us as Lisa Salvatore, Sr. Manager of Integrated Marketing at CallTrackingMetrics, walks through how to pull AEO insights, FAQ content, and real customer phrasing out of data your team is already collecting. Her colleague Brian Barranger, Sr. Account Executive III, covers what a qualified conversion actually sounds like, and how that evidence sharpens targeting, scoring, and the gaps and integration requests you route to your product team.
Reserve my Seat
CategorySEOWebinarAI Search
