Output vs Outcome vs Impact
"We shipped the new dashboard on time" gets announced as a win in almost every team, in almost every company. But shipping on time is an output claim. It says nothing about whether anyone uses it, or whether it changes anything that matters.
The three links
Output is what you produced (a feature, service, or process change). Outcome is the observable change in user or employee behavior, if the output actually works. Impact is the business, customer, or risk result that behavior change produces. Confusing 'we produced X' with 'X worked' is the single most common way teams fool themselves about progress.
A worked example
A team ships a new self-service reporting dashboard on the planned date and announces it as a success.
Output
The new reporting dashboard shipped on the planned date. A neutral fact — it carries no judgment yet.
Observable outcome
Do decision-makers actually open the dashboard and check data before meetings? This is a behavior directly observable via login or click data.
Value impact
If the outcome is real: decisions rely on more current data, and less meeting time is spent waiting for someone to pull numbers.
Notice that the output alone doesn't guarantee the outcome, and the outcome alone doesn't guarantee the impact. Each link has to actually be checked — not assumed from the one before it.
Causal humility: name the assumption
Every chain like this rests on an assumption you haven't verified yet — in the example above, that decision-makers will actually trust the new dashboard enough to drop their old reporting emails. Naming that assumption explicitly isn't a weakness in the plan. It's the one thing that tells you what to go check next.
Frequently asked questions
Isn't shipping on time itself a success?
Only at the output level. Whether it was worth building depends on the outcome and impact, neither of which shipping on time tells you.
What if I can't measure impact directly?
Most teams can't in the short term. The outcome — an observable behavior change — is the fastest honest signal you're on the path to impact.
Is it wrong to celebrate shipping something?
No — just be precise about what you're celebrating. Calling it a 'win' before you've seen the outcome is where teams fool themselves.
Now practice it
Apply the output/outcome/impact chain to a real scenario and get AI feedback — this PracticeLab is free.
Open the free PracticeLab →