GSM: Connect Your Goals to Real Metrics

Why it exists

Teams often struggle to measure success. They either track nothing, or they track dozens of 'vanity metrics' that feel good but don't connect to business outcomes. GSM imposes discipline by forcing a discussion about the underlying goal before anyone talks about a specific metric.

The mental model

Think of it like a doctor diagnosing a patient. The GOAL is 'improve patient health.' You can't measure 'health' directly. So you look for SIGNALS of health, like 'strong cardiovascular system.' Then, you measure those signals with concrete METRICS like 'resting heart rate of 60 bpm' or 'blood pressure of 120/80.' The metric is not the goal, but it is a measurable proxy for progress towards it.

How it works

You work backwards from your objective. First, state the GOAL clearly. Example: 'Increase user engagement with our new dashboard.' Second, identify the SIGNALS that would indicate you're achieving this goal. These are user actions. Example signals: 'Users are customizing the dashboard,' or 'Users are spending more time on the dashboard page.' Third, define the specific METRICS to quantify those signals. Example metrics: 'Percentage of weekly active users who save a custom view,' or 'Average session duration on the dashboard URL.'

When to use it

Use GSM at the beginning of any new project, feature launch, or quarterly planning cycle. It’s the best way to define Key Performance Indicators (KPIs) and align a team on what success looks like. It's also valuable for auditing existing dashboards to see if the metrics you track are still relevant to your current goals.

When not to use it

GSM is overkill for simple, unambiguous tasks where the metric is obvious and directly tied to the outcome, like 'Reduce server p99 latency from 200ms to 100ms.' It's a strategic framework for complex outcomes, not a simple task-management tool.

One canonical example

Let's apply GSM to an e-commerce goal. GOAL: Improve user satisfaction with the checkout process. SIGNALS of satisfaction could be a fast, frictionless process and a lack of errors. METRICS to measure these signals could be 'Average time from cart to purchase confirmation' and 'Percentage of sessions that encounter a payment validation error.' Now you have a dashboard that truly reflects the user-centric goal.