See the actual audits, reports, and content we ship — the real layouts and the real rigor, with a walkthrough film inside each block as it renders.
Illustrative where labelled — Forgeline CI, Vektrid and the example-audit brands are fictional sample clients; every figure carries its method. The live Index leaderboard in the measured strip is the one measured element on this page.
The real thing — figures, film, and the full example.
One block per deliverable: the artifact itself, a short walkthrough film, and a link to the complete example. Nothing important is hidden behind a form.
Example audits · one per tierIllustrative · four fictional brands
Share of model · “best observability tool for Kubernetes” (the free-tier snapshot; every tier deepens it)
Rationale · why this piece, this wayIllustrative · Forgeline CI
The goalReach top-3 share-of-model for “CI/CD for self-hosted Kubernetes” within 90 days.Audience: Platform engineers and DevOps leads evaluating CI/CD.
Why this angleAnswer-first + Article/FAQPage schema so engines can lift and cite it; leads with the self-hosted-Kubernetes angle where intent is highest and the field is thin.Engines disagree on CI/CD: Forgeline appears in Claude/Perplexity answers but is absent on ChatGPT/Grok — a fixable engine-specific gap. [Illustrative]
How it fits your brandMatches Forgeline's plain, technical voice; only claims the client authorized.
the goal, the sourced insights, the angle, and what to expect — beside every piece we ship
05Retainer · with every piece
The deliverable rationale
The why behind every piece: the audience and intent, the exact gap in your audit data, and the logic for the angle and format.
Watch: The deliverable rationale walkthrough (1:02) · captioned film — no audio by design · illustrative data, labelled in-film
The latest across every channel — only work that is actually published, with a real link on every card. When a feed is quiet, we say so rather than fake it.
Every number here is measured the same way — across ChatGPT, Perplexity, Claude, Gemini, and Grok — and we publish the method.
Share of model is the headline of the seven KPIs we score — with % citing your URL and % naming your brand right behind it. The full scorecard and the how live on the methodology page.
Google AI surfaces · measured separately10+ runs per enginemetric-matched 95% CIdated & re-run
Every engagement starts from the same measured picture: when AI recommends a tool in your category, who it actually names — by share-of-model across all five engines, with 95% intervals and its date. Here's a live one from our AI Visibility Index.
AI Visibility Index — CI/CD platforms · share-of-modelmeasured snapshot · 2026-07-01
#
Product
Share of model
Share
95% CI
1
GitHub Actions
21.8%
20.1–23.7
2
GitLab CI/CD
19.6%
18.0–21.4
3
CircleCI
19.1%
17.5–20.9
4
Jenkins
16.6%
15.1–18.3
5
Argo CD
6.4%
5.4–7.5
6
Buildkite
5.6%
4.7–6.7
Engines
5 of 5
Prompts
10
Answers
500
Runs/engine
10
Interval
Bootstrap 95% CI
Measured across the AI Visibility Index's 5 engines. Point-in-time; engines change. · Methodology →
Watch a measurement happen
The proof demo: an AI answer, measured — an illustrative reconstruction, labelled as such in-film.
Watch: a measurement, end to end (0:45) · captioned film — no audio by design · illustrative data, labelled in-film
Take the data with you
Your data, yours to keep — export everything to CSV or JSON anytime, no lock-in. Every deliverable — the audit, the monthly report, the QBR, the case study — ships with its data as a tidy CSV + JSON, source-labelled (what we measured vs your own numbers). And our AI Visibility Index is open data: download any category as CSV or JSON under CC BY 4.0.
Want these numbers for your category?
A free teardown shows your share of model today — measured across the AI engines, with Google's AI surfaces measured separately, plus a quick SEO read.