Why Byte Digital Built a Public Company Wiki for AI Search
Why Byte Digital Built a Public Company Wiki for AI Search
Most company websites are designed to help a prospective customer decide quickly. That is useful, but it can leave the supporting detail scattered across service pages, case studies, project files and people’s heads. We built the Byte Digital Knowledge Base to separate the commercial decision from the deeper reference work—without letting the two become competing versions of the business.
What belonged on the commercial website
The Byte Digital AI SEO and GEO service page has a commercial job. It explains who the service is for, the work Byte can provide and the sensible next step for a prospective client.
Its supporting pages cover audits, ChatGPT and Google AI search, knowledge bases, costs and working arrangements. A buyer should not need to enter a wiki to understand what they can commission.
We kept the service proposition, practical guides, client-facing case studies, calls to action and enquiry routes on the main website. Commercial pages need enough evidence to be credible, but they also need pace. If every source note, project definition and company fact sits in the sales journey, the page becomes harder to use.
What belonged in the reference library
The Byte Digital Knowledge Base has a different job. It records company identity, founder experience, service methods, sectors, projects, working arrangements and source references in a connected library.
That gave us room to publish stable facts, fuller project records, delivery methods, evidence boundaries, publication dates and source links. The distinction is deliberate: the main site helps someone decide whether Byte may be relevant; the wiki helps them inspect the basis for that decision.
One example: home-care Google Ads evidence
Byte’s UK care provider Google Ads case study is the client-facing story. It explains the campaign context, work and reported outcomes for somebody assessing the agency.
The connected home-care Google Ads wiki record gives the same evidence a reference structure. It identifies the project, defines what the reported conversion actions mean and links back to the published case study.
That connection mattered in our dated observations. On 20 October 2026, we ran six fresh ChatGPT conversations: three home-care Google Ads questions and three NHS advertising questions. The AI search observation log retains the exact prompts, visible source URLs and selected accuracy checks. One answer directly cited the wiki.
This shows that the public reference could be discovered and used in that particular ChatGPT answer. It does not show that the wiki caused all six appearances, that Gemini would make the same choice or that a citation generated a visit or qualified enquiry.
The example made the maintenance decision concrete. The commercial case study and wiki record share an evidence chain, but they serve different reading needs. A change to the underlying result definition must be reflected in both places.
Why maintaining both became worthwhile
Two public versions create work. We accepted that cost because Byte had enough distinct material to justify it: several services, long professional experience, sector knowledge, delivery methods and source-backed project records.
The library also gave repeated facts a defined home. Instead of rewriting company identity or evidence boundaries differently on every service page, we could connect readers to a maintained reference.
The implementation case study documents the architecture: one domain, a commercial section for evaluation and a reference library for deeper research. It also records delivery checks for content, canonical URLs, structured data, sitemaps and responsive pages.
The value is not the word “wiki”. It is the discipline of deciding which account is authoritative, how pages connect and who keeps them current.
Why we timestamped the evidence
Byte preserved a versioned evidence package and created an OpenTimestamps proof anchored in Bitcoin block 972195, with a recorded block time of 19 October 2026 at 19:07:41 UTC.
OpenTimestamps explains the narrow value of the method: it can prove that some data existed before a point in time. It does not prove every statement in the package is true, establish authorship or certify priority over every other company.
For us, the timestamp strengthens provenance for the preserved package. It does not establish that the knowledge base is effective, nor does it turn Byte’s qualified first-company position into an independently certified world first.
Where the early evidence stops
The October observation set concerned home-care Google Ads and NHS advertising questions in ChatGPT. It was not a Gemini test and it did not concern the phrase “AI SEO agency”.
We keep public access, search discovery, mentions, citations, recommendations, factual accuracy, referral visits and qualified enquiries as separate measures. The first items can create conditions for the later ones; they are not substitutes for them.
Our AI search measurement guide explains how to retain that separation in reporting.
When improving one service page is the better investment
A separate knowledge base is not the right first move for every business.
Improve the existing service page first when:
A well-structured service page with clear scope, named expertise, useful evidence and an honest next step can answer the buyer’s question without another content system.
Build a reference library when there is a genuine body of information that would otherwise overload commercial pages: multiple services, project records, methods, company facts, working arrangements and sources that need their own maintained structure.
More URLs are not a strategy. The additional page has to perform a distinct job.
A decision test before building a wiki
Ask:
- Which research questions cannot be answered properly by the current site?
- What source material supports the proposed reference articles?
- Which facts are stable enough to maintain publicly?
- Who will approve and update them?
- How will a reader move between reference evidence and the commercial offer?
- What would be simpler if we improved one existing page instead?
- How will access, citations, visits and enquiries be measured separately?
If the answers are vague, do not build a library yet. Fix the core pages and evidence first.
FAQs
What is a public company wiki for AI search?
It is a maintained reference library containing company facts, services, methods and evidence in pages that people and research tools can access. It should complement the commercial website rather than duplicate it.
Did the wiki make ChatGPT recommend Byte?
The October record shows six specific ChatGPT answers that recommended and cited Byte, with one answer directly citing the wiki. That does not prove the wiki caused every appearance, and it does not establish results in Gemini or for AI SEO agency questions.
Does the Bitcoin timestamp prove the wiki worked?
No. It supports that the preserved evidence package existed by the recorded time. It does not prove effectiveness, truth, authorship or historical priority on its own.
When is a service-page improvement enough?
It may be enough when the business has one clear offer, limited public evidence and no need for a separate maintained reference layer. Improve the page buyers already use before adding a library that repeats it.
Why publish both HTML and Markdown?
HTML is the normal web experience for people, browsers and search engines. Markdown provides a simpler text representation for tools that can use it. Both should come from the same maintained source so they do not drift.
Does every business need a knowledge base?
No. A knowledge base is worthwhile only when the business has enough distinct, supportable information and a real maintenance owner. Clear core pages are a higher priority for many organisations.
Oops! We could not locate your form.
Could your evidence be easier to inspect without turning the commercial site into a filing cabinet? Discuss AI SEO and GEO with Byte Digital.
Next step: Discuss AI SEO and GEO with Byte Digital
Gareth Allen is the founder and Managing Director of Byte Digital, with more than 26 years’ experience across digital marketing, design and web development. He helps businesses use paid media, websites and conversion-focused strategy to generate more qualified leads and sales. His work spans Google Ads, social advertising, campaign planning, tracking and website optimisation. With a background in design and front-end development, Gareth combines commercial marketing expertise with a practical understanding of how websites perform — from user experience and page speed to conversion tracking and landing-page improvements.