Readiness Score
A 0–100 score for how ready a website is for AI engines to read, trust and quote: crawler access, technical setup, structured data, citability and entity signals.
0–39 Not ready · 40–69 Partly ready · 70–100 AI-ready
The Cited Method
Four stages. Each ends in something you can check.
The Cited Method is how we get B2B software recommended by ChatGPT, Gemini, Claude and Perplexity: measure what AI recommends in your category, make your site readable, earn the sources AI trusts, and track the result every month.
Stage 1
Baseline
Measure what AI recommends in your category today.
- Moves
- Shortlist Rate and Readiness Score (starting point)
- Typical timeline
- About 2 weeks
We map the prompts buyers use to choose software in your category: best-of questions, alternatives, comparisons and use cases. We run each one across ChatGPT, Gemini, Claude, Perplexity and Google AI Overviews, three times per engine, and audit your website.
What you receive
- Your Shortlist Rate and Readiness Score
- Who AI recommends instead of you, prompt by prompt
- The sources AI relies on in your category
- A prioritized 90-day plan
What we don’t do: No guesswork from a single answer: every prompt is run repeatedly, because AI answers vary.
Stage 2
Readable
Fix what stops AI reading and trusting your site.
- Moves
- Readiness Score
- Typical timeline
- 4–6 weeks
We fix crawler access and rendering, add structured data, set up your entity data on Wikidata, G2, Capterra and Crunchbase, and restructure product, pricing and comparison pages so AI can quote them accurately.
What you receive
- Technical fixes and structured data across key pages
- Consistent entity data across core profiles
- Pages written to be quoted
- Readiness Score before and after
What we don’t do: No hidden text or instructions aimed at AI crawlers. Everything we publish is written for people.
Stage 3
Referenced
Earn the sources AI already trusts.
- Moves
- Source coverage and Shortlist Rate
- Typical timeline
- Ongoing, monthly
AI repeats what trusted third parties say. We earn your place in them: review platforms, comparison and best-of lists, genuine Reddit and community discussion, and trade and tier-1 press.
What you receive
- Comparison and alternatives content you approve
- Review platform presence
- Community and Reddit presence, disclosed as your brand
- Press placements
What we don’t do: No fake reviews, fake accounts or undisclosed promotion. Ever.
Stage 4
Recommended
Track the same prompts every month, and adjust.
- Moves
- Shortlist Rate
- Typical timeline
- Monthly
We rerun your prompt set every month, report changes engine by engine with the sources behind them, and move effort to whatever is holding your Shortlist Rate back.
What you receive
- Monthly Shortlist Rate report, per engine
- Comparison win rate against named competitors
- Answer accuracy: how AI describes you
- A clear plan for the next month
What we don’t do: No guaranteed placements. Nobody controls an AI answer; we guarantee the work and the measurement.
Two numbers, reported separately
Readiness is what you control and can change in weeks. The Shortlist Rate is the outcome buyers see, and it takes longer. We never present one as the other.
Shortlist Rate
The share of tracked software-buying prompts, such as “best [category] software for [use case]”, where an AI engine names your company, measured across engines and repeated runs.
0–39% Rarely cited · 40–69% Sometimes cited · 70–100% Consistently cited
How we measure the Shortlist Rate: each tracked prompt runs three times on each engine (ChatGPT, Gemini, Claude, Perplexity and Google AI Overviews), every month, from a clean session. We report per engine, with the sources behind each answer.
Principles
- No one can guarantee an AI answer. We raise how often you are recommended and report it monthly.
- No fake reviews, fake accounts or undisclosed promotion. Ever.
- No hidden text or instructions aimed at AI crawlers. Everything we publish is written for people.
- No guaranteed press. We guarantee the work and bill tier-1 placements only when they are live.
- Every number we publish has a source and a date.
See the method in practice in our results and pricing.
Last updated