The assumption that "big brand wins" doesn't hold as cleanly in AI answers as it did in SEO. Models lean on specificity, not size. This teardown is how a ~40-person challenger in a crowded B2B category took a meaningful share of the default answer away from a 4,000-person incumbent in 9 months.
The starting line
At the start of the engagement, the incumbent was named in 71% of tracked answers. Our client was at 6%. The incumbent had a 30-person marketing team. We had us.
What the incumbent was doing wrong
- Heavy PR-voice blog with no named author or specific technical detail.
- Stale G2 listing that hadn't been touched in 2 years.
- No help-doc investment. Documentation was hidden behind a login.
- Reddit presence was essentially a brand-run account that had been downvoted into oblivion.
- Technical AEO: schema missing, duplicate canonical tags on key pages.
What we did
- Wrote the 20 "Brand X vs. Incumbent" comparisons the incumbent refused to write. Honest, detailed, with tables of capability differences.
- Took help docs public. Rewrote the top 35 into answer-first pieces.
- Got the founder (former product engineer at the incumbent, incidentally) into 6 podcasts in the first 3 months.
- Rebuilt the G2 listing from scratch. Got to 40 recent reviews in 90 days.
- Built a real, signed, named-expert presence on Reddit. No corporate account. Just real people.
- Fixed the technical basics.
The curve
By month 9, the client was named in 41% of answers. The incumbent was at 58% — still ahead, but the gap had meaningfully closed. More importantly, the client was now named alongside the incumbent in most of the high-intent category queries, not just the niche ones.
Why this worked
The incumbent had stopped doing the work. Enormous marketing team, but the output was beige. Our client was specific, recent, and honest. For AI, that's what actually matters.
The takeaway for smaller brands
- Pick the 20 comparisons your incumbent refuses to write. Write them honestly. Publish them under your own brand.
- Put your most credible experts in front of AI-facing surfaces: podcasts, Reddit, specialty publications.
- Take your help docs public. The incumbent probably hasn't.
- Invest in directory listings and reviews. They still compound disproportionately.
- Don't try to out-spend the incumbent. Out-specify them.
See where your brand actually stands in AI answers.
We'll run a full custom audit before the call. You keep the report regardless.
Teardown: how a category leader got named 62% of the time
An anonymized look at the exact content, authority, and site-setup moves a category leader used to get to 62% mention rate in under 12 months.
Why AI answers became the shortlist — and what that means for your brand
The job of the top-of-funnel has quietly moved from Google's ten blue links to a single AI answer. Here's what that shift actually does to your pipeline, and what to do about it.
State of AEO — Professional services, Q2 2026
Law firms, accounting firms, boutique consultancies, healthcare practices. The AI visibility picture in services looks different than SaaS — here's how.