Recognition
Whether the systems know who you are, what you do, where you operate and who you serve.
External read, AI representation record
Before a client chooses you, many now ask an AI engine first.
The Readback is a human diagnostic read of how AI systems describe your organisation, what they get wrong, who they recommend instead, and which corrections matter first.
The visible gap
The claim and the returned answer often do not match.
The Readback captures that gap as evidence, then ranks the findings by likely commercial risk.
Illustrative record
Open finding
What the business says
“Established specialist firm, trusted by clients for complex advisory work.”
What the engine returns
> I do not have enough reliable information on that firm. For this type of work, you may also want to consider [competitor], which appears to be better established in this area.
- Finding
- Recognition weak. Competitor substituted before enquiry.
Illustrative only. A live Readback records the actual prompt, engine, date, output and source links where supplied.
Why this matters
The answer may be formed before your website is even opened.
Most organisations review their website, Google profile and search results. Fewer check the answer forming one step before that, inside the AI systems people use to shortlist, compare and explain choices.
That answer decides whether you are shortlisted, contacted, or passed over. When it is accurate and your position is clear, the people already looking for what you do are pointed towards you. When it is wrong, the enquiry goes elsewhere before you know it existed. The read exists to close that gap, so the work you should be winning is not lost at the first question.
Diagnostic scope
Five failure modes are checked before any fix is discussed.
Accuracy
Whether names, services, locations, people, prices, dates, credentials and current activity are right.
Positioning
Whether the engines understand your actual position, or flatten you into a generic category.
Substitution
Where another organisation is named instead of you, or recommended ahead of you.
Priority
Which findings are noise, and which are likely to affect trust, enquiry flow or choice.
What you receive
A dated evidence record, not a content plan.
The report identifies what is accurate, what is weak, what is wrong, and what should be corrected first.
The diagnostic report shows the external read currently being formed around your organisation.
It is designed for practical judgement, not performance theatre. You can use it internally, hand it to your web, comms or SEO team, or ask for a separate correction brief.
Prompt log
Engine outputs
Competitor mentions
Accuracy notes
Risk order
Correction priorities
What the read typically finds
The work you are losing never shows up as a missed call.
Most organisations discover at least one of these. Some find all six.
Not recognised
The engines draw a blank or return only the most basic facts. No description of what the firm does, who it serves or what makes it different. A competitor fills the gap.
Named incorrectly
A wrong principal, a former partner, a previous name, a defunct location. The engines are carrying a version of the firm that no longer exists.
Substituted
A competitor is named ahead of you, or in place of you entirely, when a prospective client asks which firm to use. You are not in the conversation.
Flattened
The engines know you exist but reduce you to a generic category. Your actual position, specialism or differentiator does not reach the answer.
Stale
Prices, services, team members, accreditations or credentials that changed months or years ago are still being reported as current. The engines have not caught up.
Contradicted
What the engines say about you conflicts with what you say about yourself. The claim and the returned answer do not match, and the engine's version reaches the client first.
The reader
A person reads it, not a dashboard.
The Readback is run by Chris Luffingham, and every read is conducted personally. The prompts can be automated. The reading cannot. Knowing whether an engine has flattened your position, named a competitor in your place, or quietly gone out of date takes someone who can weigh the answer against what is actually true, and judge which findings carry commercial risk and which are noise.
The method was proven in the charity and not-for-profit sector, where institutional funders use AI to form a view of an applicant before any meeting takes place. The subject never sees that read being formed, and the decision is large. Accuracy and source discipline were not optional there. That is the standard each commercial read is now held to.
If you have received a report, Chris prepared it.
Next step
See the read before others act on it.
Request a diagnostic read, or email Chris directly if you already have a report and want to talk through the findings.