Developing
Opinion that promotes the author's "behavior-led search" approach; examples are illustrative and some points are speculative.
In 60 seconds
- AI search lets people express their full context, needs and trade-offs, not just keywords; examples include a family hotel search and a running shoe "up to $150".
- AI memory of past conversations may shape answers even when that context is not in the prompt, though memory and personalisation vary by platform and settings.
- Prompt tracking has value, but a standardised prompt shows only one version of possible answers.
- Search insight can inform business strategy, such as niche positioning and situational content.
Impact Layer
- What changed
- A practitioner argues prompts reveal richer customer intent than keywords.
- Why it matters
- If answers vary by user context, one tracked prompt can mislead teams about visibility.
- Who is affected
- SEO, GEO and content strategists.
- Martech stack area
- SEO & Search, Content
- Stack impact
- Track several phrasings of the same need with different contexts, and use the prompts customers write as research.
- Business impact
- Low A contributed opinion piece on MarTech (owned by Semrush) that cites no surveys, studies or data; the author runs an agency and promotes her own approach.
- Related technologies
- What to watch
- Studies of how memory changes AI search answers.
Becky Simms, founder of Reflect Digital, argues in a MarTech column published October 20, 2026 that AI search is turning prompts into customer research. The page says the opinions are her own.
The main claims
- AI search lets people express their full context, needs and trade-offs, not just keywords. Her illustrations are a Tenerife hotel search for a family of two adults and two young children, and a buyer of running shoes willing to spend “up to $150”.
- AI memory of past conversations may shape answers even when that context is not in the prompt.
- Prompt tracking has value, but a standardised prompt shows only one version of possible answers.
- Search insight can inform business strategy, such as niche positioning and situational content.
Caveats
- The article cites no surveys, studies or quantitative data; its claims rest on the author’s observations and illustrative examples.
- Memory and personalisation vary by platform and user settings, so they will not apply to every AI search experience.
- Some points are speculative: “It seems likely that as AI continues to mature, it, too, will know.”
- It promotes the author’s “behavior-led search” approach.
Our read
The practical idea, to read the prompts your customers write, does not depend on the rest. Test it with real queries from your own support and sales channels.
Sources
- MarTech: AI search is turning prompts into customer research (opens in a new tab)
Source for the column's claims and caveats.