How Do We Know Question

Derive key results from objectives by asking 'How do we know we succeeded?'

Christina Wodtke
The ultimate guide to OKRs

How Do We Know Question

"I love, 'how do we know?' That's how you get to outcomes. So, what does it mean to have a vision for a product we believe will be successful and meet our mission? Well, what would it be? How are we going to figure this out?" - Christina Wodtke

What It Is

"How do we know?" is the key question that bridges inspirational objectives to measurable key results. When you have an objective like "Create a vision for our new product," asking "how do we know we succeeded?" forces you to identify the concrete evidence that would prove success.

This simple question prevents two common OKR failures: vague objectives that can't be measured, and key results that are just tasks rather than outcomes.

How It Works

The Process:

  1. Start with the Objective - A qualitative, inspiring goal for the quarter

  2. Ask "How do we know?" - What evidence would prove we achieved this?

  3. Brainstorm measurements - Spend 10 minutes generating every possible way to measure success

  4. Select 2-3 key results - Choose a mix of:

    • Hardcore numbers (users, revenue, engagement)
    • Quality signals (NPS, satisfaction, delight)
    • Business impact (often with a dollar sign)

Example:

Objective: "We have a vision for this game that will drive us forward."

How do we know?

  • User testing shows strong interest → X% of test users want to play more
  • Technical feasibility proven → Working prototype demonstrates core mechanic
  • Market potential validated → Landing page generates X signups

How to Apply It

  1. Write your objective first - Make it inspiring but concrete enough to evaluate

  2. Gather your team for brainstorming - Don't shortcut this step

  3. Set a 10-minute timer - This forces you past the obvious ideas:

    • First few minutes: obvious metrics everyone thinks of
    • Middle period: you feel stuck and run out of ideas
    • Last few minutes: weird, creative ideas emerge (often the best ones)

  4. Challenge each idea - For each potential measurement, ask:

    • Is this an outcome or a task?
    • Can we actually measure this?
    • Would hitting this number mean we truly succeeded?

  5. Pick triangulating metrics - Choose key results that together give confidence. If one is high and another is low, you'd question success.

  6. Accept imperfection - "If you don't know, you just set it and you'll know by the end of the quarter whether you were stupid or not."

When to Use It

Use this question when:

  • Converting strategy into quarterly OKRs
  • Objectives feel fluffy or unmeasurable
  • Key results keep turning into task lists
  • Team disagrees on what success looks like
  • You're stuck between many possible metrics

The insight: Everything can be measured to some degree. The question isn't whether you can measure perfectly, but whether you can learn something useful. Start measuring imperfectly and improve over time.

Common Failure Modes

Task contamination: "Ship the feature by March" is a task, not an outcome. Ask: What happens after we ship it? That's the outcome.

Vanity metrics: "Get 1M impressions" might not indicate success. Ask: What does that impressions number mean for our actual goal?

Single metrics: One number can be gamed. Use 2-3 key results that triangulate—if they all move together, you're probably succeeding.

Source

  • Guest: Christina Wodtke
  • Episode: "The ultimate guide to OKRs"
  • Key Discussion: (00:19:35) - Using "how do we know?" to derive key results
  • YouTube: Watch on YouTube

Related Frameworks

Back to Execution