Define what each metric can establish
A sampled mention shows an output under recorded conditions. A referred visit shows a visit attributed by the available system. An enquiry shows a customer action, but its full discovery path may be unknown. Do not multiply these into a claimed revenue figure without supporting evidence. Record what is unavailable rather than filling gaps with an invented model of certainty.
Connect reporting to a practical decision
Choose what the business will do differently after the review. An inaccurate description may justify a factual clarification; relevant referred visits may justify closer examination of the destination page. A changing mention count alone may not justify a new publishing programme. Keep the question set, observation period and measurement changes visible so later comparisons remain interpretable.
A practical check before you buy
Take one reporting period and classify every headline number as an answer sample, visit, enquiry or attributed sale. Ask what evidence connects adjacent stages and mark unknown connections explicitly.
| Decision | What to consider | What to check |
|---|---|---|
| Answer sample | Observed output | Keep prompt and date |
| Referred visit | Measured arrival | Document attribution limits |
| Business enquiry | Actual customer action | Check quality and follow-up |
An illustrative business scenario
A UK consultancy with a long sales cycle can keep AI referral observations separate from later enquiries whose full research journey is unknown.
This is an illustrative scenario, not a claim about an ImbaHub client or a promised result.
Discuss the right scope with ImbaHub
ImbaHub builds fast static websites with Astro, with editing through Imba CMS and Imba Builder. Start with the customer task described in this guide and ask which content, integrations and ongoing responsibilities your project needs. Agree the scope before comparing the investment.