Lesson 7 of 7 in Run an Agency or Multi-Client Practice
Why this matters
Clients can absorb bad news. What damages the relationship is discovering later that a result was overstated, that reach was presented as revenue, or that Unknown was reported as a conclusion. Signal & Science keeps those distinctions visible: Observed versus Unknown in the diagnosis, recorded results with their source, pending results marked as pending. Your job on the call is to carry the same discipline into your own words, so the client's Learn step rests on what actually happened and your recommendation keeps its credibility.
When to use it
Every client review, and any time you share a diagnosis, a briefing, or a result, including informal updates by email or chat.
Before you start
- The client's Marketing Plan open, with the target, recent tasks, and recorded results in view.
- Open decisions and missing outcomes cleared, or a clear list of which are still open and why.
- The Client review agenda or your own equivalent.
Do this
- 1Start with the outcome the plan is driving toward (baseline, current, target), or say plainly that no target is set yet.
- 2Report what was observed in the diagnosis's own terms. The app defines Observed as cited evidence the scan found, and Unknown as an area it could not verify, which does not mean the business is failing.
- 3Name what remains Unknown and the next step that would verify it. Unknown is not zero and not a poor grade.
- 4List what shipped, task by task.
- 5Give what the result shows: the measured value, where it came from, the dates it covers, and whether it is task-linked or workspace-wide (see Separate task-linked metrics from workspace metrics). Report pending results as pending.
- 6State what it does not show. Reach, views, and impressions are distribution evidence, not business impact, and a change observed after the work is not proof the work caused it.
- 7Close with the decision for next week (keep, simplify, skip, or change) and who owns it. Record it in the plan so the next briefing reflects it.
What you should see
The client asks sharper questions instead of nodding along, and disagreements surface while they are cheap to resolve. Your notes from the call map to entries in the workspace: a decision, a recorded result, or a new Research source.
How to judge it
After the call, the client could repeat back what is known, what is Unknown, and what was decided, and nothing you said claims more than the recorded results support.
What it does not mean
Honesty about uncertainty is not hedging, and it is not withholding a view. Unknown with a named next verification is a plan; Unknown with a shrug is not. Human judgment still sets direction: give your recommendation, and be clear about the evidence behind it and what would change your mind (see Interpret met, partial, missed, and inconclusive results).
Try it in Signal & Science
Open Clients, choose the client you meet next, and prepare what is known, what is Unknown, and what needs a decision.
Open in Signal & Science