AI search isn’t on the horizon; it’s here, and it’s already picking winners. After watching Marketing Against the Grain break down how brands rise or sink in answer engines, my view is simple: optimize for answers, not just pages. If your plan is still built around classic SEO, you’re handing share to smarter rivals.
AI Search Demands A New Playbook
Kipp Bodnar and Kieran Flanagan set the stage, then Asia Frost and Beria Amiel showed how AI systems choose what to recommend. The message hit hard: your website is only a slice of what drives AI results. Reddit threads, media coverage, YouTube, LinkedIn, and peer blogs steer large chunks of these recommendations. Keep polishing your site if you want, but don’t expect it to carry the load.
“You’re in a constant daily battle with your competitors for the share of voice and the recommendations across these AI search engines.”
As someone who’s shipped products and content for decades, I’ve watched shifts like this reward fast movers. The winners now are the brands that earn citations from sources AI trusts, and learn which prompts matter most to their buyers.
What The Pros Showed
Asia and Beria walked through prompt tracking and daily monitoring that shows when and how a brand gets mentioned. The point wasn’t theory; it was performance. They highlighted three ideas marketers need to accept:
“Your website is only doing a small part of the work, and you’re probably spending way more time on it than on all those other places.”
“We send these prompts to the answer engines every single day… your visibility will adjust slightly every single day.”
“In terms of all the prompts that mentioned a brand, Dell had 52.8% of those mentions whereas Lenovo had 42%.”
That example also showed why sentiment matters. Flooding the feed with mentions doesn’t help if the tone is negative. The team mapped which channels shaped opinions most. For the case they showed, peer content drove ~55%, PR ~26%, and the brand’s own site only ~4%. That should wake up any leader still fixated on homepage tweaks.
They even surfaced the content formats AI likes to cite—such as listicles—and tied recommendations to measurable lift on specific prompts. That last mile—proving an action raised visibility—separates busywork from momentum.
My Take: Focus On Prompts, Proof, And Places
This shift rewards precision over volume. Keywords are a blunt tool. Prompts map to real buyer intent by product and persona. You need to see which questions surface your brand, which name your rivals, and which ignore you.
I also agree with their daily cadence. AI answers move more than classic search results. If you don’t check often, you’ll miss swings that your competitors catch.
Here’s the simple plan I recommend as a marketer and builder who’s lived through many platform shifts:
- Map prompts to products and personas, then track share of voice and sentiment daily.
- Publish where AI “listens”: Reddit, YouTube, LinkedIn, peer blogs, and earned media.
- Create content in citation-friendly formats (yes, smart listicles), not just opinion posts.
- Earn mentions from credible peers; don’t rely on your own site to do the heavy lifting.
- Test actions against prompt-level visibility and keep what moves the needle.
For crypto, social, and online business, this is even sharper. AI engines weigh trust signals. Community-backed explainers, transparent FAQs, and third-party breakdowns beat hype. I’d seed thoughtful comparisons, credible pros-and-cons, and clear pricing pages that others will cite. That’s how you win answers without gaming the system.
Addressing Doubts
Worried this favors big brands? It can—but only if you concede the field. Smaller teams can still punch up by targeting specific prompts, owning underserved questions, and earning citations from niche experts. Another fear is volatility. That’s real, which is why a daily loop of measure–ship–measure keeps you ahead while slower teams wait for quarterly slides.
The Bottom Line
Stop treating AI search like old SEO. The game is prompt-level visibility, credible citations, and steady proof of lift. If your dashboard can’t tell you where you gained or lost mention share today, you’re flying blind.
Start this week: pick one product, define five buyer prompts, publish two citation-friendly pieces off-site, and measure changes in mention share and sentiment. Do it again next week. Then again.
AI is already choosing. Make sure it chooses you.
