The Signal
What should a sales intelligence report include? An annotated example
A useful sales intelligence report says which companies deserve attention, what changed, why that could matter to your offer, and what to do next. It carries dated sources, keeps evidence separate from interpretation, names what is still unknown, and gives the next action an owner. Here is the structure, with a worked example you can take apart.
By Alfredo Silva 11 min read
Imagine an alert telling you a manufacturer is opening a second site.
The announcement may be accurate. The company may fit your target market. You still have to decide whether there is a reason to contact anybody, and what you could credibly say.
A useful report helps you make that decision. Just as importantly, it tells you when the available information is too thin to justify one.
Start with the decision the report has to support
A report for a chemical distributor should look nothing like one for an operations consultant. The same expansion announcement raises questions about supply, distribution agreements, staffing or management process depending on who is reading it.
So define the brief before collecting anything:
- Your offer. The problem you help customers solve.
- Your target. Company types, geography and the roles that would own the issue.
- Your exclusions. What makes an account unsuitable however attractive it looks.
- Your capacity. How many accounts you can genuinely investigate or approach.
- Your decision. Which companies to pursue, which to monitor, which to set aside for want of evidence.
That gives the researcher a reason to include something. It also gives them permission to leave things out, which is the half most briefs forget.
What belongs in the report
| Component | What it should tell the reader |
|---|---|
| Scope and review date | Which market, offer and period this covers. |
| Decision summary | What deserves action, what needs watching, what lacks evidence. |
| Account fit | Why this company matches the brief, including the exclusions checked. |
| Observed change | What happened, when it happened, and when it was found. |
| Source record | Where each observation came from and how to check it. |
| Interpretation | Why the observation could matter commercially. |
| Unknowns and contrary evidence | What could weaken or overturn the reading. |
| Relevant role | Who might own the issue, and whether the contact was verified. |
| Next action | What to do, who owns it, when it gets reviewed. |
| Outcome history | What happened after the recommendation, and what that taught you. |
None of this needs a page per field. A short summary and a few well-supported account briefs are usually enough.
An annotated example
Everything in this example is invented. "Example Manufacturer A", its announcements, its dates and its research notes were written to demonstrate the format. It is not a real prospect, not a customer result, and not an extract from a delivered SONAR™ report.
The commercial brief
An independent operations consultant helps manufacturers make planning and handovers consistent across sites. They want to investigate a small number of Dutch manufacturers adding a second location. Companies already using a competing consultancy for this work are excluded. There is capacity to investigate three accounts this week.
Why this belongs. Without the offer and the exclusions, an expansion announcement has no commercial meaning at all.
The decision at a glance
| Field | Entry |
|---|---|
| Account | Example Manufacturer A |
| Reviewed | 7 October 2026 |
| Fit | Dutch manufacturer planning a second production location. Current advisory relationships unverified. |
| Decision | Pursue a short verification step. Outreach depends on what it establishes. |
| Commercial hypothesis | A second site may create work around planning and operational handovers. |
| Main uncertainty | The business may already have the processes and the support it needs. |
| Next owner | Consultant |
| Next review | 9 October 2026, or sooner if the check resolves the uncertainty. |
Why this belongs. The reader sees the recommendation and its limit in the same glance. Pursue names a specific action. It does not mean the company is ready to buy.
1. Observed: what the evidence actually says
| ID | Observation | Source and dates |
|---|---|---|
| S1 | The company announced plans to open a second production location in January 2027. | Invented company announcement, published 29 September 2026, reviewed 7 October. |
| S2 | An operations vacancy mentions coordinating production planning across both locations. | Invented company vacancy, published 2 October 2026, reviewed 7 October. |
Swipe the table sideways →
A real report would carry the source URLs and the supporting passages. There are none here because these documents do not exist.
Why this belongs. Publication date, event date and review date answer three different questions. An old announcement found today is not a new event.
2. Interpretation: what it might mean
Together, a planned site opening and that vacancy suggest coordination across locations is worth investigating, and that sits inside the consultant's work.
They do not establish that planning is currently failing, that anyone is overwhelmed, or that budget exists for outside help. There is also a perfectly plausible alternative: the company may be hiring precisely in order to build that capability in-house, in which case the vacancy is evidence against the opportunity rather than for it.
Why this belongs. Interpretation turns information into a hypothesis. Naming the alternative reading is what lets the reader judge how much weight to give it.
3. Check: what to resolve before any outreach
- Verify the opening is still planned and the vacancy still live.
- Check for a publicly named implementation or advisory partner.
- Identify the person responsible for coordinating the two sites.
The exclusion check stays open until there is enough information. Failing to find a named consultant on a website does not prove none is involved.
Why this belongs. A report should reduce uncertainty in a defined way. "Research further" gives the owner nothing to do.
4. Action: what to do with the result
If the company still fits and the check supports an approach, contact the verified operations lead with a question grounded in the announcement:
I saw the announcement about your second production site. How are you approaching planning and handovers between the two locations?
That is an illustrative opening, conditional on the verification above. It asks about a situation without claiming to know about an internal problem.
If the support is already in place, close the item for this offer. If the opening is postponed, move it to monitoring with a trigger. If the announcement cannot be verified at all, mark it insufficient evidence and move on.
Why this belongs. A recommendation should say how to proceed and what would change the decision.
Use decision labels that mean something
The labels matter less than their definitions. This set works:
| Decision | Meaning | Required next step |
|---|---|---|
| Pursue | Enough fit and evidence to justify a specified action. | Name the action, the owner and the deadline. Say whether it is verification or outreach. |
| Monitor | Relevant, but action depends on timing or another development. | Name the event to watch, or a review date. |
| Insufficient evidence | Too weak, stale or incomplete to support a recommendation. | Identify what is missing and whether getting it is worth the effort. |
Swipe the table sideways →
An account that fails an explicit fit criterion can simply be excluded with the reason recorded. Keep that separate from an account you do not yet understand: they look similar in a list and they are completely different decisions.
Separate confidence in the facts from confidence in the opportunity
An official announcement is strong evidence that an expansion was announced. It is much weaker evidence that the business needs your particular service. Assess those two separately, because a report that scores them as one number hides the weaker half.
A numerical score can organise a queue when its criteria are written down. Without that, 92 out of 100 creates a precision the reader cannot evaluate, and it should never quietly become a 92% chance of winning the customer. There is more on what a score is actually fitted to in lead scoring.
Then ask what would change the assessment. A newly announced adviser, a postponed project, or a reply saying the issue is already solved should all move the recommendation.
Close the loop after somebody acts
Put the response and the outcome back on the same account record. Was the proposed person responsible for the issue? Did the situation hold up? Did they recognise the need? Did a conversation, a meeting or an opportunity follow? What should change in future selection?
Keep the stages distinct, because this is where reporting quietly inflates. An accepted invitation is not an attended meeting, and an opportunity is not a signed customer.
A worked example from our own published work: in the EOS Implementer engagement, SONAR™ scanned 766 construction businesses and produced 20 verified owners, ranked by the moment their business was in, each with a verified direct line and a specific reason to call. Jim made the calls, and two owners who had never met him agreed to join his next session.
Those are three numbers measuring three different things, and research of this kind is answerable for the middle one. Whether the data was right, whether it surfaced the right people, and whether it gave him something worth saying is what you can judge it on. The conversation after that belongs to the person making the call, and a report that quietly claims credit for it is the same inflation this section is about.
A reusable account brief
Copy this for each account worth reviewing. It is plain text on purpose, so it drops into a CRM note, a document or a spreadsheet without bringing formatting with it.
ACCOUNT BRIEF ============= Account and fit : company, criteria met, exclusions checked Reviewed : date, researcher Observed : specific events, with dates Sources : links, publication dates, supporting passages Interpretation : why this could matter to your offer Alternative reading : another plausible explanation of the same evidence Unknowns : what you cannot yet establish Decision : pursue / monitor / insufficient evidence / excluded Next action : a specific verification, outreach or monitoring step Owner and timing : responsible person, deadline or trigger Outcome : response, stage reached, what it taught you
Before handing a report over, ask one question of it: could the recipient explain why this account is here, check the evidence themselves, and take the next step without another briefing? If so, it is doing its job.
How this relates to SONAR™
SONAR™ is the engine that watches a market for change and connects it to commercial opportunity, moving from signal to situation to action. The structure above is a way to evaluate that kind of work, ours included. The format can vary. What should not vary is that a reader can trace any recommendation back to its evidence and see what is still unknown.
Buying this kind of work from someone else? Question 2 of the twelve is about making a provider walk one example from observation to approach.