Detect hidden assumptions in reported dashboard numbers

Surface filtering tricks, omitted baselines, and date cutoffs hiding behind pasted dashboard metrics. Use this before presenting figures to executive leadership.

You getApparent Conclusion and hidden Assumptions and Blind Spots

Prompt

Prompt
Act as an executive data auditor. Evaluate the pasted dashboard metrics, summaries, or report notes below.
- Expose unstated assumptions in how the data was filtered, aggregated, or sampled.
- Identify what missing denominators, survivor biases, or cutoffs could distort the true business picture.
- Provide direct, non-defensive questions to ask the data owner to confirm accuracy.
Output format:
1. Apparent Conclusion (1 sentence on what the numbers claim)
2. Hidden Assumptions and Blind Spots (numbered list with specific risks)
3. Questions for the Analyst (3-5 exact questions to ask the creator)
Here is the dashboard data:

Example output

Example for: sales demo close rate reported at 42% on an executive summary

  1. Apparent Conclusion
    The report claims that the sales team achieved an exceptional 42% conversion rate from completed product demonstrations to signed contracts last month.
  2. Hidden Assumptions and Blind Spots
  • Survivor bias in meeting definitions: Only prospects who completed the full 45-minute call were likely counted, ignoring no-shows, reschedules, and early drop-offs.
  • Pipeline cutoff timing: Deals may be counted as won in the month closed while attributing them back to meetings from older months, skewing the actual cohort performance.
  • Excluded deal sizes: Low-tier self-service signups might be combined with enterprise sales demos to artificially inflate the win count.
  • Single-segment skew: One high-performing rep or a single batch of enterprise renewals could be masking low win rates across standard inbound opportunities.
  1. Questions for the Analyst
  2. Are deals counted in the 42% tied strictly to demos held this calendar month, or is this won deals divided by total demos regardless of vintage?
  3. How are demo no-shows and prospects disqualified in the first five minutes categorized in this calculation?
  4. What is the median close rate by individual account executive across this same period?
  5. Does this close rate metric include existing customer expansions and renewals, or strictly new logos?
Written by AI from this prompt. Your results will vary.

Paste the metrics summary, KPI cards, or report commentary right below the prompt in ChatGPT, Claude, or Gemini. Follow up by asking what charts or split views would reveal the underlying distribution.

Make it repeatable

Keep using this prompt without pasting it in every time.

Keep exploring

More in Data Analysis