Write a stakeholder update

Output

What a PM had to fix

1. 1Hypothesis stated as factReframe it as a hypothesisStart again

   Source: MetricsMobile only: −5.2%.

   What we’d changeLead with the strongest comparison: mobile fell 5.2% with no navigation change, so the navigation isn't established as a cause. Say what you'll check and when.

2. 2Invented evidenceVerify or remove the claimSubstantial rework

   Source: IncidentAbout 30% of teams use single sign-on.

   What we’d change30% of teams use SSO. That doesn't mean all of them were trying to get in during those two hours.

3. 3Hypothesis stated as factReframe it as a hypothesisQuick edit

   What we’d changeBeta teams may differ from everyone else. Their small drop doesn't show a manageable learning curve.

4. 4Hypothesis stated as factReframe it as a hypothesisQuick edit

   Source: Metricswhich we put down to the start of the school year.

   What we’d changeLast year's cause was an attribution, not a finding. Say 'which we put down to'.

Check by check

Got wrong · 6

• Uses the supplied evidence correctlyIt introduces causal statements such as 'directly impacting mid-week engagement' and 'new navigation is a factor' that are not in the supplied context or derivable by arithmetic.
• Addresses the actual decisionIt does not say what result or condition would change the answer; it only says it will monitor and share a breakdown.
• Identifies material uncertaintyIt does not name the specific unknowns that could change the decision or tie the monitoring to what would settle the call.
• Avoids unsupported claimsIt presents causal interpretations such as the SSO outage directly impacting engagement and navigation being a factor as established fact, though the evidence only suggests them.
• Separates effect from explanationIt lists SSO outage, navigation friction, and seasonality as 'what happened' rather than clearly labelling them as hypotheses with supporting or contradicting evidence.
• Owns the problemIt reports the SSO outage but does not say what the team has already done about it or recommend a way forward for that specific problem.

Got right · 4

• Respects explicit constraintsIt is a channel reply addressed to Lena and is under 200 words.
• Produces the required deliverableThe reply is in the right channel format, addressed to Lena, under 200 words, and usable as a status update.
• Answers the question askedThe first lines answer the question by saying the new navigation is a factor but not the whole story.
• Ready for the reader as deliveredIt is written for Lena, uses no placeholders, and all dates and owners are taken from the material or clearly proposed.

Claims the judge couldn’t find in the brief

• The new navigation is a factor, though not the whole story.
• The SSO outage directly impacted mid-week engagement.
• This suggests the learning curve is real but manageable once users adjust.

Output

Check by check

Got wrong · 1

• Owns the problemIt proposes future analysis and fixes but does not state any actions the team has already taken or offer trade-offs.

Got right · 9

• Uses the supplied evidence correctlyAll factual numbers, dates, and quotes are drawn accurately from the supplied context, with interpretations clearly hedged.
• Addresses the actual decisionThe output commits early that navigation is unlikely to be the whole story and says tomorrow's analysis will show whether rollback is supported.
• Respects explicit constraintsThe reply is under 200 words, addressed to Lena, and fits a leadership-channel update.
• Identifies material uncertaintyIt names the non-representative beta cohort, unproven seasonality, and SSO outage as unknowns and outlines the analysis that will resolve them.
• Avoids unsupported claimsCauses are labelled as possible ('may be contributing', 'plausible', 'suggesting') rather than asserted as fact.
• Produces the required deliverableIt is a complete channel reply with an answer, evidence, and next steps, within the requested length.
• Answers the question askedThe first line directly addresses whether the new navigation caused the drop.
• Separates effect from explanationEach candidate explanation is framed as a hypothesis with supporting or contradicting evidence (mobile unchanged, beta low, seasonality, SSO).
• Ready for the reader as deliveredIt is written entirely for Lena with no placeholders or unexplained internal labels.