Build a brand that search and AI can identify, describe accurately and cite.

Organizations for which being described wrongly, or not at all, has a measurable cost.

• Premium and high-trust brands whose value depends on being understood precisely, not broadly.
• Scale-ups and technology companies entering a category that AI assistants already describe through older competitors.
• Investment firms assessed by founders, limited partners and co-investors before any conversation takes place.
• Professional-services firms whose expertise sits with partners whose names are better known than the firm's.
• Institutions with a public mandate and a long, fragmented digital record.
• Organizations in complex or reputationally sensitive markets, where one inaccurate summary travels far.

Two of these profiles are examined on a page of their own: professional-services firms and institutions. All are listed in Audiences.

Eight workstreams, scoped to the organization's situation. This is the full range of the work; what the starting fee covers is set out under Investment below. The discipline itself is explained in what AI reputation is, and how it is managed.

• Brand narrative and category definition The category you belong to, the claim you can defend, and the vocabulary buyers and machines should associate with you.
• Organizational entity architecture One coherent definition of the organization, its brands, products and locations, expressed in structured data such as schema.org Organization markup and consistent across every profile.
• Executive-to-brand alignment The public record of the people who lead the company supports what the company says about itself. See reputation and authority for leaders.
• Proof, research and expertise assets Evidence that others can cite: data, methodology, documented results, original analysis.
• Third-party authority and citation strategy The independent sources that should describe the organization, and the legitimate reasons they would.
• Technical SEO, AEO and GEO The implementation that makes accurate information easy to find, extract and cite. See how the disciplines of AI visibility fit together.
• AI search baseline and competitive reading A dated baseline of how the brand is described, cited and compared with named competitors. Tracking over time is the optional monthly engagement, AI Reputation Intelligence.
• Governance across markets and languages Who owns the record, how updates propagate, and how versions in different languages stay consistent.

Each workstream maps to a layer of The Authority Architecture System.

1. Ask for the diagnosis before the plan. A serious proposal starts from what assistants say about you today, with the prompts, engines and dates recorded.
2. Identify your actual bottleneck. Missing third-party sources, pages that are hard to crawl, an unclear narrative and absent proof assets call for different skills.
3. Check how results will be measured. Mention rate, citation share and accuracy are observable. A single blended "AI score" with no method behind it is not.
4. Reject guarantees. No provider controls a model's output. Guaranteed citations or recommendations are a warning sign.
5. Ask who writes and who implements. Entity and structured-data work requires technical execution, not only strategy documents.
6. Clarify confidentiality. Decide in advance what may be published about the engagement.

Brand Authority Build — Starting at $28,000 USD

• Billing One-off fixed-fee engagement.
• Indicative duration 12 to 16 weeks.
• Scope at the starting price One organization, its main entity and one reference leader; one language and one market; entity architecture, narrative, evidence library, up to eight reference pages or assets, and technical optimization.
• What is not included Press relations, media placements, video production, native translation, legal advice, data purchases, third-party tools, out-of-scope development and travel are not included unless agreed in writing. A second market, a second language, several leaders, an active crisis, a large volume of content or a multi-entity architecture are quoted separately.

All amounts are in US dollars (USD), excluding taxes and third-party costs. These fees do not include any guarantee of ranking, citation, recommendation by an AI system, removal of legitimate content or media outcome.

See every engagement, fee and term

A Build above $15,000 USD can be paid in three installments: 50% at launch, 30% at interim validation and 20% on delivery. If a Build is commissioned within 30 days of the presentation of the assessment findings, the validated assessment work is deducted from the Build fee, up to $4,500 USD.

How do AI assistants decide what to say about a brand?

They combine what the model learned in training with sources retrieved by web search when the question is asked. They favor information that is consistent across sources, corroborated by independent publications and easy to extract. A brand described differently on each channel gives them nothing stable to repeat.

Why does ChatGPT recommend our competitors and not us?

Possible reasons include competitors appearing in the comparison and review sources the assistant retrieves, an entity that is clearer and better corroborated than yours, or content of your own that does not answer the question in a form that can be quoted. The diagnostic page sets out how to check which one applies.

How long does it take to see results?

There is no fixed timeline. Assistants that search the web can reflect new or corrected sources once those are indexed and retrieved. Without live web access, an answer may rely on information captured earlier; timing and coverage vary by provider and model. Work on third-party sources also depends on other publishers' schedules.

Can anyone guarantee citations in ChatGPT or Google AI Overviews?

No. Answers vary with the question, the user, the engine and the day, and no provider controls them. What can be built and measured is the quality, consistency and corroboration of the evidence these systems draw on.

Is this the same as working with a GEO agency?

Not quite. A GEO agency typically produces and distributes content to earn citations. Authority work starts earlier: it defines the entity, the narrative and the proof that content should carry. Many organizations need both, in that order.

Do you work across several markets and languages?

Yes. The work is delivered in English and French, and the governance layer addresses how an organization's record stays consistent across countries and languages.