Company research worksheet

Record a company claim, its original source and the checks needed before relying on it.

Your working copy

Fill in the fields you can answer and mark missing information as unknown. Entries are not sent to B&F. Save a TXT copy for work you want to keep.

Claim and intended useWrite one sentence. Which entity, period and definition does it cover, and which decision does it affect?Original source and locationRecord publisher, title, date/version, public link or permitted record reference, exact passage and access date.What is actually reportedRecord only the information supported by the passage, including scope and units.Assumptions and judgmentWhat connects the reported information to your conclusion? Label the inference.Alternative and source dependenceWhat else could explain the observation? Do apparent confirmations share one source?Evidence that could change the claimName the missing record or fact precisely; avoid a generic “needs verification”.Revised claim and reviewWhat can you say now? Who checks the remaining issue, and when must the claim be revisited?

Download blank template

Export text

Instructions

Make the claim testable
Write one precise sentence with entity or group, geography, period and definition. Name its intended use: screening, an interview question or a transaction decision.
Record: One bounded claim and its use.
Open the underlying record
Find the original filing, company publication or other relevant record. Capture publisher, date, version and exact page, table or passage. Consider direct knowledge and the author’s interest. Record when the original is inaccessible.
Record: A traceable record, not a search snippet or unsupported AI citation.
Separate information and inference
Write what the source reports, then the assumptions needed for the conclusion. A website supports “the company describes itself this way”; it may not confirm present capacity, customer mix or performance.
Record: Separate reported information, assumptions and judgment.
Check alternatives and dependence
Look for an explanation that would change the conclusion. Acquisitions, group boundaries or prices can affect revenue. Articles repeating one release are not independent confirmations.
Record: An alternative explanation and evidence that would distinguish it.
Decide what the evidence permits
Keep, qualify or remove the claim. Name the missing fact, appropriate reviewer and refresh trigger. Confidence in the evidence is not a numerical probability of a future event.
Record: A corrected claim, next check and limit of use.

Worked example

Worked example / Fictional

Fictional worked example. All businesses, documents and amounts are invented for learning; no client result is claimed.

Revenue growth does not establish organic growth

Claim to check: Company A’s core business grew organically by 20% in 2025.
Fictional source record 1: Invented consolidated annual report, FY2025, revenue table p.12: reported revenue is CZK 100 million in 2024 and CZK 120 million in 2025.
Fictional source record 2: The same invented report, acquisition note p.24: a distribution business was acquired during 2025. The extract does not quantify its revenue contribution.
Fictional source record 3: An invented trade article repeats the company press release’s 20% figure; it adds no separate evidence on like-for-like growth.
What the numbers support: The reported increase is (120 − 100) / 100 = 20%. This verifies the arithmetic on the teaching inputs, not the accuracy of any real accounts.
Unsupported assumption: The original claim assumes the full increase comes from the same underlying business. The acquisition means this is not established.
Alternative explanation: Acquired revenue could explain part or all of the increase. Price, volume or other changes could also matter; their effects are not supplied.
Source dependence: The trade article repeats the release. It is not an independent check of organic growth.
Corrected claim: The fictional FY2025 report shows 20% higher reported consolidated revenue. The acquisition contribution is unavailable, so organic growth cannot be determined from these inputs.
Next evidence and reviewer: Obtain a revenue bridge or comparable disclosure separating the acquisition contribution and clarifying reporting scope and basis. A transaction use requires interpretation by an appropriately qualified financial reviewer.
Current decision: Do not mark the organic-growth criterion as met. Keep it unresolved until the missing evidence is reviewed. This card does not assess acquisition suitability.

The arithmetic can be correct while the business conclusion is unsupported. Improve the sentence before collecting more copies of the same claim.

Sources and limits

Original B&F application. These sources inform the method; they do not validate this tool or establish results for your business.

Our application of analytical standards, not a validated scoring system or guarantee of accuracy.
Public records can be incomplete or dated. An absent public record does not prove an activity, risk or relationship is absent.
This is a research aid, not a valuation, audit, legal opinion or full financial due diligence.
A second AI answer or a repeated article is not independent verification. Review must match the subject and consequences of error.

Analytical standard
Intelligence Community Directive 203: Analytic Standards
Describe source quality; separate reported information, assumptions and judgments; consider alternatives and evidence that would change the conclusion.
Passages examined and limitations
D.6.e.(1)–(7), PDF pp.4–6; information, assumptions and judgments on p.5.
Developed for US government analysis. Our company claim card is an adaptation, not an official instrument or certification.
Analytical quality guidance
The AQuA Book
Agree the question and constraints; check both correct execution and fitness for the decision; record data, assumptions, uncertainty and review.
Passages examined and limitations
Sections 6.1–6.4, 8.2.1–8.2.2 and 8.3–8.4.
A quality framework, not a trial of these guides. Checks should match the intended use and consequences of error.

Linked publications keep their own terms. Free access does not grant permission to republish them. Reuse the original B&F worksheet.

Your working copy

Instructions

Make the claim testable

Open the underlying record

Separate information and inference

Check alternatives and dependence

Decide what the evidence permits

Worked example

Sources and limits

Intelligence Community Directive 203: Analytic Standards

The AQuA Book