All resources

Research / Review guide / 3 min read

Read a shortlist: evidence, inference and what is still unknown.

Make lead recommendations easy to inspect and easier to improve.

THE EVIDENCE LADDERSeparate what you know from what you infer.
01 / OBSERVED“Wholesale PV supply”

Public company description

02 / INFERREDPotential distribution fit

Relevant to this mission

03 / STILL UNKNOWNInterest in a new supplier

A question, not a fact

Illustrative framework. Use verified information from your own business.

A useful lead record makes it possible to disagree with the recommendation. Reviewers need to see what was observed, why it matters to the mission and which parts still require confirmation.

01

Read the evidence before the recommendation.

Open the company website and check that the cited page supports the stated business role. A search snippet is a starting point, not a complete company profile. Record when the information was reviewed and whether the page belongs to the business being evaluated.

02

Test the inference against the mission.

Ask whether the evidence supports this particular audience and offer. A company may be a strong fit for one mission and a poor fit for another. Keep business fit separate from contact availability and from any numerical ranking used to sort the list.

03

Give feedback that changes the next decision.

When rejecting a lead, explain the mismatch in terms the team can reuse: customer segment, geography, product focus or commercial role. When accepting it, retain the unresolved questions. A kept lead means it is worth progressing; it does not establish purchase intent.

IN PRACTICE

Make a recommendation auditable.

LESS USEFUL

Great lead. High score. Likely to be interested.

MORE ACTIONABLE

The website describes wholesale PV supply. This supports a possible distributor fit; additional-supplier interest remains unknown.

An explanation should survive the removal of the score. A reviewer should still understand the business reason and what must be checked.

Record elementWhat good looks like
Observed factSpecific statement with a source.
Mission inferenceWhy that fact matters to this offer.
UncertaintyA clear question, not hidden confidence.
TRY THIS WITH YOUR TEAM

Choose one accepted and one rejected lead. Hide their scores and read the explanations. If you cannot tell why the decisions differ, rewrite the reasoning around role, customer segment, geography or product relevance.

MAKE IT YOURS

A worksheet for your team.

Use this as a starting point. Replace the fields with verified information about your own business.

Lead and mission:
Source URL and review date:
Observed fact:
Why this could fit:
What remains unknown:
Contact route and confidence:
Keep / investigate / exclude:
Feedback for future research:
Download text file

Before you move on.

  • Source ownership and date are visible.
  • Facts and inference are separate.
  • Fit and contactability are evaluated separately.
  • Feedback names the actual mismatch.

Editorial guidance for planning and review. Examples are illustrative. Confirm your workspace’s available controls and configuration before live execution.