Services

One prerequisite, three delivery lines.

Every engagement starts with a baseline diagnostic, included in scope and not charged separately. It is not a product you can buy on its own: it requires meeting the people who run the business and going through product material, pricing logic and the buying chain — and its conclusions set the standard for everything that follows. The three delivery lines are combined according to those conclusions and cannot be bought without the diagnostic.

Prerequisite · included in the engagement

Baseline diagnostic

This is not an entry product you can try out cheaply. It requires meeting the people who run the business, reviewing product material, customer mix, pricing logic and the buying chain — then designing the probe set around how buyers actually ask. Its conclusions set the standard for the whole engagement: fact sheet, corpus map, site structure and every later round are measured against what the diagnostic fixed.

Business and product brief — product lines, customer mix, buying chain, and the questions we could not resolve
Probe question set — five purchase-intent layers, each question with its decision rule
5 × N capture matrix — per-cell answer text, citation sources and timestamps
Four baseline metrics — awareness, accuracy, visibility, top-3 mention rate
Factual deviation list — every incorrect claim, with the correct fact and its evidence
Competitive snapshot — which other brands appear for the same questions, and how often

Line 02 · scheduled from diagnostic findings

Entity and corpus building

Before retrieving facts, an engine must decide which entity the answer is about. If your brand cannot be resolved reliably — or gets merged with a same-named company — every downstream content effort is discounted.

Brand fact sheet — name, aliases, founding date, location, credentials, product lines, key figures
Corpus map — target question × expected source × existing asset, with gaps flagged
Owned-media rewrite — definition-style paragraphs, citable data blocks, Q&A structure
Earned-media plan — third-party sources ranked by credibility, with their inclusion criteria

Line 03 · scheduled from diagnostic findings

Site GEO technical retrofit

Upgrading the site from “written for humans” to machine-readable first, still pleasant for humans. This does not change your business — only how your content can be read.

Crawler accessEntity declarationsQ&A structureDefinition-style paragraphsllms.txtCitable data blocksNo-JS readabilityMultilingual annotation

Line 04 · Quarterly

Quarterly re-measurement & iteration

GEO is not a one-off project. Engines update models, reweight retrieval sources, and competitors are working on the same problem. Optimization without continuous measurement is driving with your eyes closed.

Four rounds per quarter, with the question set held constant so that variance is attributable.

Metric values and round-over-round deltas
Cell-level diff: which cells moved from “not mentioned” to “mentioned”, and which regressed
Attribution analysis, including explicitly unattributable changes
Next-round priorities ranked by cells affected ÷ cost

Pricing

Why we don’t publish rates

The diagnostic is not charged separately; it is part of the engagement. Pricing then depends on what the diagnostic finds — two companies in the same industry can differ twofold in product lines, customer mix and buying-chain complexity. Site hardening also depends on your stack and page count. A single public price would overcharge small clients or undercharge large ones; both are dishonest.

The process: diagnostic call → confirm question set and current state → one-page quote stating scope, deliverables, timeline and what is not included → you confirm, we start. No hidden line items.

→ Start a diagnostic call