NorthStar Behaviour

FRAMEFOUNDATIONS

One sharp action that guides every decision.

What
Define the single user behaviour your team must drive. The tool turns vague goals into one clear, observable action with a success signal, time frame, and business link.

When
Run it at the start of any product, feature, or strategy initiative, and review it every quarter or after a major pivot.

Why
Without a North Star behaviour, teams chase vanity metrics, build random features, and debate aimlessly. A single, measurable action aligns product, design, data, and leadership, keeps experiments on target, and links human behaviour directly to revenue.

Common questions:

“Is this just OKRs in disguise?”
→ No. OKRs often focus on outputs (features launched, sessions increased). The North‑Star Behaviour is about one observable user action that predicts success.
“Why only one behaviour? Aren’t there lots we want?”
→ Sure, but if everything’s important, nothing is. The North‑Star filters noise. Other behaviours might matter later, but you need a clear target first.
“How is this different from a KPI or business metric?”
→ A KPI measures business performance. This canvas defines the human action that drives that performance. ARR is not a behaviour. “Setting up an auto-save goal” is.

STEPS

Step 1: Set the Table

Brief the group on the mission:
“We will define one user action that makes or breaks this product/service.”

Invite Product, Design, Data, Marketing, Decision‑maker.
Lay the five Mindset cards in view and read them aloud.
Explain to everyone that you are focusing specifically on the things users do (user action) vs talking about behaviour. 'Behaviour' as a word trips people up. It's used to represent a wide array of things.

Step 2: State the Product/Service Goal

Write a single line that explains the value you want to create.
Example: “Help Gen Z users build an emergency fund.”
Keep it under twelve words.

Step 3: Brain‑Dump Actions

Each person writes as many user actions as they can, one per sticky.
Format: verb + object – “connects social”, “sets auto‑save”, “shares streak”.

Step 4: Spot Real Behaviours

Use the Behaviour vs Non‑Behaviour Cheat Sheet (below) to quickly discard or reformat the non-behaviours.

Cross out feelings, metrics, and passive states.
What stays must be observable, countable, repeatable.

Step 5: Map Impact and Certainty

Stick surviving actions on a quick 2×2 Leverage Grid:

Impact (low→high behaviour shift) on Y‑axis.
Certainty (we can influence) on X‑axis.
Circle the top‑right cluster.

Step 6: Vote for One Behaviour

Silent dot‑vote on the circled actions.
Choose based on these three tests:

Product wins if it happens.
Product fails if it doesn’t.
Team can influence it within 90 days.

Step 7: Craft the Behaviour Statement

Fill in this sentence:
“[Who] [does what] within [time frame].”
Example: “New UK user enables Auto‑Save within 24 hours of sign‑up.”

A note on the who: Pull from your existing personas or segments. But keep in mind that influence travels, and there might be a better who hidden beneath.

Step 8: Define the Success Signal

Locate or create the exact event in your analytics (API call, log, tag).
Write it: “Event: autosave_enabled = true.”
If no event exists, add a ticket to create it.

Step 9: Capture Baseline

Ask Data to pull yesterday’s number. Or last month's overall performance.
If you can't or it's missing, plan a 48‑hour sample to estimate. Write it down.
For clarity later, write the user stories that data represents.

Step 10: Link to Value

Use the Behaviour‑to‑£ Calculator:
(Baseline uplift × Average Revenue per User) – cost to build.
Write the headline: “Each Auto‑Save user is worth £32 more per year.”

Step 11: Set the Target Shift (3 min)

From X to Y in the next quarter. Keep bold yet believable.
Example: “Move from 4 % to 16 % within 90 days.”

Step 12: Run the Red‑Flag Filters

Visibility: Can we see the behaviour daily?
Remit: Do we have control over the levers?
Ethics: Is influencing fair and transparent?
If any answer is “No”, revisit Steps 6‑8.

Step 13: Commit in Public

Paste the final North Star line at the top of your shared doc.
Each team member types their name beside it. Public commitment locks it in.

Step 14: Schedule the Review

Book a 90‑day calendar invite titled “North Star Check‑In.”
If data moves faster, bring it forward.

Expected Outcome

One sharp statement everyone can quote:

“A first‑time user enables Auto‑Save within 24 hours of sign‑up, raising their 90‑day retention by 16%. With a value of £32 per user per year”

This drives every map, idea, prototype, and test that follows.

Resources

Behaviour vs Non‑Behaviour Cheat Sheet

Use this when deciding if a sticky note is a real action.

A real behaviour checks three boxes:

You can see it happen (observable).
You can count it (binary or numeric).
A user can repeat it (not a one‑off state).

Quick test:
Say, “I can watch a real person do X right now.” If you can’t picture it, rewrite.

Data‑Signal Glossary

Three ready‑to‑copy event schemas and tips to track your North Star.

Naming tips

Use past‑tense verbs: enabled, created, connected.
Keep it lower_case with underscores for readability.
Include a timestamp and the user identifier every time.

How to find or create the right event

Check analytics docs: Search for verbs that match your target action.
Inspect the code: Look for the event fired in the function that handles the behaviour.
Use network tools: Trigger the action in staging, open the browser network tab, and watch for event calls.
No event? Log it: Ask engineering to add one. Keep it lightweight, just the essentials above.
Validate: Perform the action twice in a test account and confirm two events land in your data store.

Good practice

Track the first and the repeat occurrence separately if habit matters.
Store the source (mobile, web) to spot channel bias.
Add a human‑readable description in your analytics dictionary so new team members know exactly what the event means.