Working with Stakeholders¶

Discuss goals¶

Stakeholder requests are usually tied to a bigger project or goal. When they ask for something, take the opportunity to learn more.

• Start a discussion
• Ask about the kind of results they want
• A quick chat about goals can set expectations and plan next steps

Feel empowered to say "no"¶

A marketing director with a "high-priority" project needs data by tomorrow morning to back their hypothesis. Sometimes:

• Their hypothesis isn't fully formed
• You have a better approach
• The work will take longer than estimated

Push back when you need to. Saying "no" with reasoning earns more trust than agreeing then under-delivering.

Plan for the unexpected¶

Before starting:

• List potential roadblocks
• When discussing timelines, build in buffer for problem-solving
• Communicate risks early, not when they hit

Know your project¶

• Track discussions over email/reports
• Be ready to answer how aspects matter to the org
• Understand how your project connects to company-wide goals
• The deeper your context, the more effective your analysis

Start with words and visuals¶

Analysts and stakeholders interpret things differently while assuming alignment. This illusion of agreement sends projects back-and-forth.

• Start with a description and quick visual of what you're conveying
• Stakeholders prefer different formats — words or pictures
• Iterate from there
• Faster agreement = faster analysis = faster feedback = faster learning

Communicate often¶

• Stakeholders want regular updates
• Share notes on milestones, setbacks, changes
• Use a change log — chronological list of project modifications
• Make updates pull-able, not just push: link to a doc anyone can check

Five questions before any project¶

• What kind of results are needed?
• Who will be informed?
• Am I answering the question being asked?
• How quickly does a decision need to be made?
• Where's the data going to come from?

Find the right blend¶

Data + business knowledge + gut instinct.

• Rush projects can't always wait on perfect data
• Gut works when you've seen the pattern before
• Too much gut → bias
• Blend carefully

Common stakeholder types¶

References¶

• HBR — How to Run a Data-Driven Stakeholder Meeting
• Storytelling with Data
• Atlassian — Stakeholder communication