Case studies
Self-directed practice work carried out to build GEO skill, supporting an SEO team in Pakistan. All figures come from the tracking platform's own API snapshots, taken before and after the work, on a 30-day rolling window. Brands are anonymised by sector and region; the named version with full monthly reports is available on request.
Custom embroidery & workwear · Orange County, California
2 → 6 Aug 2026
0 %→≈ 9 %AI visibility, 30 d
0→≈ 49brand mentions
≈ 3.9average rank
≈ 77sentiment
The brand started completely invisible: zero mentions across every tracked question. The tracked set was the problem — full of national "best custom embroidery in the US" questions owned by CustomInk and 4imprint.
- Read the answer transcripts — citations existed, but only on Laguna Hills and Orange County questions.
- Paused 24 national and zero-yield questions, kept the 7 local ones already producing brand citations.
- Added 12 new questions built from Search Console — the exact local terms the site already earned impressions for.
- Set a rule I now follow everywhere: pause, never delete — re-adding costs answer-run credits.
Commercial print shop · Irvine / Orange County, California
1 → 6 Aug 2026
Technical AI readiness
1 %→≈ 3 %AI visibility, 30 d
4→≈ 49brand mentions
6.3→≈ 4.0average rank
94AI readiness · 16/17
The hardest case. At the start the shop appeared in 1% of answers with four mentions, while Vistaprint held 19% and Staples 13%. You do not beat Vistaprint on "business cards" — so the strategy moved to local commercial intent, and the site was rebuilt to be machine-readable.
- HowTo schema in JSON-LD on the home, same-day and file-upload pages — the one failing readiness check, fixed.
- Markdown version of every page, served by content negotiation, so an agent requesting text/markdown gets clean text.
- Agent discovery — /.well-known/agent.json, MCP server card, Link headers, and a WebMCP script exposing business info and services.
- Question set rebuilt: 23 local active, 34 national paused, twelve taken from Search Console.
Two benchmark brands · print, California & salon/spa, Ontario
Aug 2026 baseline
25 %visibility · print brand
92mentions · 4th of 51
9 %visibility · salon
3.8average rank · salon
Measured to establish what "good" looks like in a local market. The print brand sits fourth out of 51 cited entities in its region, ahead of every independent rival; the salon holds 9% where the strongest local spa has 18%. Both produced the content brief: the page types models quote, and the directory profiles they build answers from.
What these numbers do and don't prove
Part of each jump comes from changing which questions are measured — dropping national ones a small business can never win. That is a deliberate strategy decision, not a trick, but it means the percentage is not a pure like-for-like comparison.
The clean number is mentions: 0 → 49 and 4 → 49 are real citations of a real brand name inside real AI answers, and they did not exist before. Every figure traces back to a dated API snapshot stored with the project.
How the work is run
- Monthly performance report per brand — visibility, mentions, rank, sentiment, calls and website clicks
- Question inventory with IDs, state and the reason each one moved
- Dated raw JSON snapshots so any number can be checked afterwards
- Written project rules: what must not be broken
- New tracked questions sourced from real 90-day Search Console data