AI assistants are becoming a new layer between customers and companies. maesee shows whether your business is visible, understood, trusted and recommendable using observed model behaviour and evidence-based scoring.
The shift
For two decades, businesses optimised for search engines, websites, apps and marketplaces. A new layer is now emerging between customers and businesses: the AI assistant.
A growing number of customer journeys begin with questions like:
If AI systems cannot confidently find, understand and recommend your business, you may lose future demand before customers ever reach your website.
Prompt evidence
maesee tests the answers AI systems give when customers ask real discovery, comparison and recommendation questions.
Extracted evidence
Illustrative example showing the type of evidence maesee captures from observed model responses.
The framework
The structured assessment framework maesee runs. It measures observed AI behaviour across five stages, each scored against documented, evidence-based criteria.
The product
A fast, evidence-based diagnostic that answers the first and most important question: how visible and recommendable is your business to AI systems today?
Define the diagnostic scope
We agree the market, competitors, customer scenarios and priority use cases before testing begins.
Run a calibrated AI test set
We test carefully selected discovery, comparison and recommendation scenarios using a repeatable methodology tailored to your market.
Capture the model evidence
Every AI response is recorded, reviewed and converted into structured evidence for analysis
Apply rule-based scoring
Scores are calculated against documented DURTA criteria, based on observed model behaviour rather than subjective opinion.
Deliver findings and next steps
You receive a clear, plain-language report, practical recommendations and a walkthrough of the evidence behind the score.
Methodology
maesee scores observable AI behaviour. Every finding is backed by captured model responses. Every score can be traced to documented criteria.
We test AI models directly and capture their actual responses. Scores are based on what AI does, not what it might do.
Every response is structured against consistent criteria before scoring. Evidence is traceable and reproducible.
Scores are calculated by applying documented rules to extracted evidence. No subjective AI opinion is involved.
Limitation: Scores reflect observed model response behaviour under controlled, repeatable testing. They are a diagnostic indicator and AI visibility proxy, not a definitive measure across all AI systems. Every report includes a methodology note covering the models tested, test set version, scoring version, audit date and confidence levels.
Example output
The Snapshot produces a structured AI Distribution Readiness Report. Below is an illustrative example.
Who it is for
maesee is designed for organisations that rely on digital acquisition, discovery and customer trust.
If new customers find you through search, comparison and recommendation today, AI-mediated discovery will matter to you tomorrow.
Outcomes
maesee shows you where you stand, why, and what to do next.