Every example below is invented. The sample uses no client identity, client data or client result.
What should the first page tell you?
You should not have to turn a dashboard into a business decision.
The first page should answer:
- What work was completed?
- What changed in the evidence?
- What did not change?
- What does that mean for the business?
- What should happen next?
The example below uses invented values and generic assets. It shows the reporting structure without implying a real result.
A one-page decision summary
| Question | Invented example |
|---|---|
| What changed? | The approved service page became eligible for indexing and began receiving non-branded impressions |
| What did not change? | Qualified enquiries showed no reliable movement in the same period |
| Which work supports the change? | Page rewrite, internal-link repair and indexing check in the work record |
| What does it mean? | Discovery improved; commercial effect remains unknown |
| What happens next? | Improve and test the action path on the reviewed page |
| Who owns it? | Named page owner and operations owner |
| When is the next review? | After the agreed window and enough comparable enquiries |
A credible report can use the word “unknown.” The evidence should decide the wording.
Record the work before claiming a result.
| Work item | Asset | Reason | Completion evidence | Status |
|---|---|---|---|---|
| Clarify the service answer | Example service page | Help the buyer understand fit and next action | Dated page version | Complete; impact not established |
| Repair the internal route | Example hub and service page | Make the page reachable from its parent | Crawl record | Complete; indexing response pending |
| Test the enquiry handoff | Example request form | Confirm the request reaches the right owner | Test submission record | Test passed; real enquiry quality unknown |
“SEO completed” is not a useful row. The record should name the asset, reason, date and completion evidence.
Keep work, visibility and revenue in separate rows.
| Record | What it establishes | What it does not establish |
|---|---|---|
| Completed work | The agreed task was finished | Search visibility or business impact |
| Indexing or profile eligibility | The asset can participate under checked conditions | A ranking or buyer response |
| Search, Maps or AI visibility | The platform reported or the test observed exposure | A visit or recommendation |
| Website visit | A tracked session reached the site | Qualification |
| Onsite action | A configured call, form, booking, demo or purchase event fired | A valid opportunity or sale |
| Qualified action | The business judged the action against its rule | A completed commercial result |
| Commercial outcome | The operating system recorded the result | Sole causation by one page or answer |
This prevents a visibility improvement from being reported as revenue before the business records support it.
Make every important number explain itself.
A useful metric needs:
- the source and exact report;
- the date range and comparison period;
- the property, page, profile or market in scope;
- the denominator and exclusions;
- the extraction date;
- any tracking or method change;
- the known limitation.
“Organic traffic increased” leaves too much hidden. A reviewable statement names the source, page group, comparison and classification method. It still says that clicks do not prove qualified business.
Show what did not move.
A flat result should not disappear inside a blended score.
The approved page gained impressions, but qualified enquiry volume did not show reliable movement during the review period. The evidence does not support a commercial-impact claim yet.
That sentence is more useful than a positive headline built from the first metric alone.
Use the right final record for each industry.
| Industry | Qualified action | Commercial record |
|---|---|---|
| Home Services | Valid request in the served area | Booked estimate, scheduled job or completed work |
| Healthcare | Appropriate request routed to the correct service | Scheduled appropriate appointment |
| Legal | Prospective matter meeting approved criteria | Accepted consultation or signed matter where compliant |
| B2B SaaS | Account meeting the qualification rule | Qualified opportunity and pipeline record |
| eCommerce | Valid order after approved exclusions | Completed order, return-adjusted sale and repeat purchase |
One “leads” column cannot describe all five businesses honestly.
How should AI answers be reported?
An AI observation should include the exact question, tool, date, known location or account context, answer, visible sources and sample limit.
One answer shows what happened in that test. It does not establish permanent platform behavior, audience size or commercial value.
See Measuring AI Answers Honestly for the full method.
End with a decision someone owns.
| Field | Invented example |
|---|---|
| Decision | Improve the action path on the approved high-fit service page |
| Reason | Discovery improved; qualified actions did not |
| Owner | Named page owner and operations owner |
| Expected signal | More valid completions without lower qualification |
| Review point | After the agreed observation window |
| Reversal condition | Stop or roll back if quality falls or the change creates a barrier |
A recommendation without an owner and review point is easy to forget. One without a reversal condition is easy to defend forever.
Who is responsible for the report?
Mindflow delivers the engagement. Tools may collect exports, crawl pages and organize evidence. Specialists may review technical, legal, clinical or privacy questions when the scope needs them.
I decide what the findings mean, write or approve the client-facing recommendation and remain responsible for it. I do not claim to perform every specialist task.
What is excluded from this sample?
The sample contains no client identity, testimonial, private account data or real performance result.
Do not send patient information, legal-client information, personal call records, customer lists or order-level identity through a general web form.
If you have a report you cannot act on, the Free Visibility Check can identify the first visible reporting or conversion problem. A paid assessment may be needed when the answer depends on private data.