Matching
A query is compared with passages of text, by shared terms and by meaning. The passage that scores highest is retrieved.
BM25(q, d) and cos(q, p)
How it works
Measure the retrieval. Change what moves it. Measure again.
Search engines and AI systems decide what to retrieve with scoring functions. Nobody outside those companies sees their exact functions, so we calculate the standard retrieval measures they are built on, for your business and your competitors, check them against what AI systems actually answer, and change the information that moves them.
Retrieval
What decides who is retrieved.
Four mechanisms. Each can be estimated for your pages and your competitors' with the standard measure behind it.
Understanding
Systems recognise entities: organisations, services, places, people. Facts stated explicitly and consistently are understood. Facts left to inference often are not.
attributes stated / attributes expected
Structure
Links between pages tell systems which topics connect and which pages matter. Authority flows through the link graph.
PR(p) = (1 − d)/N + d · Σ PR(i)/L(i)
Answering
AI systems build answers from retrieved passages and name the companies those passages support. If your passage is not retrieved, you are not named.
answer share = k / n
The method
Diagnose. Implement. Validate.
-
01
Diagnose
We build the query set for one service category, index your pages and your competitors' at passage level, and score them. The Retrieval Opportunity Report sets out every gap, its cause and its fix, ranked by commercial value.
-
02
Implement
We carry out the roadmap: entity and ontology work, page restructuring and internal link architecture. Each change is specified against the measure it should move.
-
03
Validate
We re-run the same queries against the same measures and report what moved, query by query. If a score did not move, we say so and why.
Diagnosis
Implementation
Entity & Ontology Engineering
Retrieval-Focused Page Architecture
Internal Link & Topic Architecture
Validation
How it starts
From application to roadmap.
-
01
Apply
Answer six short questions about your business. We read every application.
-
02
15-minute consultation
Free. We show you two or three verified retrieval weaknesses in your market.
-
03
Retrieval Opportunity Report
A measured diagnosis of one service category, and a prioritised roadmap.
-
04
Implementation
The roadmap carried out, then re-tested against the same measures.
What you can hold us to
Every engagement is scoped, measured and re-tested.
- A fixed scope and named deliverables, agreed before work starts
- Every change specified against the measure it should move
- Results reported against the original measures, query by query
- Scores from our own models, stated as estimates. What AI systems answer is observed, not modelled
- No guaranteed rankings or enquiries: nobody controls how Google or an AI system ranks
Free 15-minute consultation
Find out where your business stands.
- How search and AI systems currently interpret your business
- Where competitors are better understood than you
- Specific retrieval weaknesses, checked before we speak
- Whether there is work worth doing, and what it is worth to you
Apply for a free 15-minute consultation
We reply within one business day, and usually within the hour.