Communicating Insights

Turning Analysis Into Action

Analysis has no value until someone acts on it. Communication is where analysis becomes impact.

The Communication Challenge

Analysts vs. Decision-Makers

Analysts think in: Methodology, data quality, statistical significance, nuance.

Decision-makers need: Clear answers, confidence level, recommendations, implications.

Bridging the Gap

Your job isn't to present analysis. It's to enable decisions.

Know Your Audience

Executive Audience

Wants: Bottom line, implications, recommendations Time: Minutes, not hours Format: Executive summary, headlines, key visuals Detail: On request only

Technical Audience

Wants: Methodology, data sources, assumptions Time: Will read deeper if needed Format: Structured report with appendices Detail: Full detail available

Operational Audience

Wants: What to do, specific actions Time: Focused reading Format: Clear recommendations with supporting evidence Detail: Enough to understand "why"

Structure Your Communication

The Pyramid Principle

Start with the answer. Support with evidence. Provide detail as needed.

  1. Lead with the conclusion
  2. Provide supporting arguments
  3. Back up with data/evidence
  4. Include detail in appendix

The SCQA Framework

  • Situation: Context the audience knows
  • Complication: What changed or what's the problem
  • Question: What this raises
  • Answer: Your insight

One Page Summary

Force yourself to fit findings on one page:

  • Headline finding (1 sentence)
  • Key supporting points (3-5 bullets)
  • One key visual
  • Recommendation
  • Next steps

Writing About Data

Lead with Insight, Not Data

Bad: "The data shows that Q4 had 12,456 transactions averaging $47.32."

Good: "Q4 transactions were 15% below target, driven by a 23% decline in new customer purchases."

Be Specific

Vague: "Sales increased significantly."

Specific: "Sales increased 18% year-over-year."

Explain "So What"

After stating a finding, explain why it matters.

Finding: "Customer acquisition cost increased 25%." So what: "At this rate, acquiring new customers will become unprofitable within two quarters."

Acknowledge Uncertainty

Be honest about confidence level:

  • "We're confident that..."
  • "The data suggests..."
  • "Further investigation is needed to confirm..."

Presenting Visuals

One Chart, One Message

Each chart should make one clear point. Label it clearly.

Tell the Chart's Story

Don't just show the chart. Explain:

  • What it shows
  • What to notice
  • What it means

Avoid Chart Crimes

  • Don't truncate axes to exaggerate
  • Don't use 3D or fancy effects
  • Don't overcrowd
  • Don't hide important information

Handling Questions

Anticipate Questions

Think about what stakeholders will ask:

  • What about [other factor]?
  • How confident are you?
  • What should we do?
  • What happens if we're wrong?

Prepare Backup Detail

Have deeper analysis ready to discuss if asked. Don't present it all upfront.

Admit What You Don't Know

"I don't know, but I can find out" is better than making things up.

From Insights to Recommendations

Don't Just Describe — Prescribe

Analysis that ends with "here's what we found" is incomplete.

Move to "here's what we should do."

SMART Recommendations

  • Specific: What exactly to do
  • Measurable: How to know if it worked
  • Achievable: Realistic given constraints
  • Relevant: Connected to the analysis
  • Time-bound: When to do it

Risk and Trade-offs

Good recommendations acknowledge:

  • What could go wrong
  • What you're trading off
  • What assumptions underlie the recommendation

Documentation

For Future Reference

Document your analysis so others (including future you) can:

  • Understand what was done
  • Reproduce results
  • Build on the work

Include

  • Question addressed
  • Data used and sources
  • Methodology
  • Key findings
  • Limitations
  • Recommendations

AI Prompt: Creating Communication

Help me communicate my analysis findings.

Key findings:
1. [Finding 1]
2. [Finding 2]
3. [Finding 3]

Audience: [Who will receive this]
Context: [How this will be used]
Format: [Email, presentation, report]

Please create:
1. A compelling opening that states the main insight
2. Structured supporting points
3. Clear recommendations
4. Anticipated questions and responses

AI Prompt: Simplifying Technical Content

Help me explain this technical finding to a non-technical audience.

Technical finding: [Your statistical or analytical result]
Audience: [Who needs to understand]

Translate this into:
1. Plain language explanation
2. Relevant analogy or comparison
3. Why it matters to them
4. What action it implies

What's Next

Let's apply these skills to real scenarios.

Next chapter: Common analysis scenarios.

Using AI for AnalysisCommon Analysis Scenarios