Free Tool · Noma

GEO Readiness Scorecard

Answer 12 questions to find out how ready your brand is to appear in ChatGPT, Gemini, Perplexity, and Google AI Overviews.

0 of 12 answered
1. Does your website use structured data markup (schema.org)?
2. Do your key pages have clear question-answering H2/H3 headings?
3. Do you have FAQ sections on your product or category pages?
4. Is your content written in a clear, factual, first-person brand voice?
5. Do you publish original data, research, or case studies?
6. Are your pages cited or linked by authoritative external sites?
7. Does your About page list team credentials or company background?
8. Is your brand mentioned in any industry publications?
9. Do product pages clearly explain what the product does in one short paragraph?
10. Do you use FAQPage, HowTo, or Product schema on relevant pages?
11. Is your key content updated at least once per quarter?
12. Does your brand have a Wikipedia, Crunchbase, or Wikidata entry?

Common questions

What is GEO, and how is it different from SEO?

Generative Engine Optimisation is the work of getting a brand cited inside AI-generated answers rather than ranked in a list of blue links. SEO competes for a position on a results page a person then chooses from. GEO competes to be the source a model quotes when it answers instead. The overlap is real — both reward clear structure and credible sourcing — but ranking first for a keyword does not guarantee being the brand ChatGPT names.

How is the score calculated?

Twelve questions, each answered on a scale, normalised to a score out of 100. The bands are 0–30 Not Visible, 31–50 Emerging, 51–69 Developing, 70–85 Visible and 86–100 AI-Ready. Because you answer the questions yourself it is a structured self-audit rather than a crawl of your site — its value is in showing which of the twelve signals you are missing, not in the number.

Which of the twelve matters most?

Knowledge graph presence — a Wikipedia, Wikidata or Crunchbase entry — and third-party mentions in industry publications tend to carry the most weight, because they are the hardest to fake and the models lean on them to decide a brand is real. Schema markup and clear question-answering headings matter too and are far quicker to fix. Start with the structural ones you control, then work on the citations you do not.

Why does structured data matter so much to AI engines?

Because it removes guesswork. Schema tells a crawler what a page is, what the product does and who published it, in a form that needs no interpretation. FAQPage, HowTo and Product schema in particular map directly onto the shapes AI answers take. Without it a model has to infer all of that from prose, and inference is where a brand gets described inaccurately or skipped.

How long does it take to show up in AI answers?

Longer than an SEO change, and less predictably. Structural fixes such as schema and headings can be picked up within weeks of a recrawl. The signals that actually move brand recognition — third-party mentions, original data others cite, knowledge graph presence — take months, because they depend on other people publishing. Anyone promising AI visibility in a fortnight is selling something.

Do the different engines want different things?

Broadly the same signals, weighted differently. Perplexity leans hard on fresh, citable web sources and shows its references, so recency and linkable pages matter more. ChatGPT blends its training data with browsing, which makes long-standing brand mentions count for more. Google AI Overviews draw on the same index as search, so classic SEO strength carries over most directly there. Fixing the twelve items here helps with all of them.

Can I pay to appear in AI answers?

Not today, in any of the major engines, in the way you can buy a search ad. Citation is earned through the signals in this scorecard. That will not stay true forever — advertising is arriving in AI interfaces — but the organic citation itself is not currently purchasable, which is precisely why the structural work is worth doing now rather than later.

What do I do after scoring?

The scorecard lists your weakest answers as improvements, in priority order. Take the structural ones first because they are entirely within your control: schema on key pages, clear question-shaped headings, FAQ sections, and a one-paragraph plain description of what your product does. Then measure. Tracking how the engines actually describe your brand over time is what Noma does, and it is the difference between doing GEO work and knowing whether it worked.