Learn the three responses.
See when Stay, Adjust, or Shift is justified, and what evidence could change the conclusion.
Compare the responses ↓START HERE
You can execute a strategy well while the assumptions beneath it become less reliable. Polaris Shift helps teams and organizations decide what to hold steady, what to adjust, and when to change direction.
Start small: Understand the idea, try a fictional decision, then leave with a brief you can download and discuss. No account required. Explore five practical questions ↓
A guiding outcome used to make trade-offs. It can guide a team, a function, a business unit, or a whole organization. A metric measures progress toward it; a project or quarterly target is not automatically a North Star.
A North Star provides continuity while conditions change. Changing the route can preserve that outcome. Reconsidering the destination requires evidence; staying the course is a credible decision.
Don’t just follow the North Star. Know when to move it.
YOUR FIRST VISIT
See when Stay, Adjust, or Shift is justified, and what evidence could change the conclusion.
Compare the responses ↓Enter a fictional leadership scenario. Choose a direction, reveal evidence, and reconsider the trade-off.
Try a 3-minute scenario →Start with three short prompts and prepare a brief for a strategy conversation. Keep unknowns visible.
Start with three questions →Before scaling an AI pilot. When customer behavior diverges from a healthy dashboard. When a team disagrees about whether a new capability changes the strategy.
A routine execution problem may need better execution. A compelling demo may warrant a test. Neither automatically calls for a new goal.
BEFORE YOU CHANGE DIRECTION
Start with the question your team is stuck on. These are working explanations of the Polaris framework, not a diagnostic score or a substitute for evidence about your organization.
Imagine a mapmaker. Helping people navigate could remain its purpose. Making the best paper maps might have become its governing product goal. Live navigation would invite leaders to test whether that goal still represents what travelers need, and whether the company can create value in a different way. This is a fictional illustration, not proof that every mapmaker should change course.
Changing a measure can improve how you observe the same outcome. Changing the governing outcome changes the trade-offs you are willing to make. Name which level is under discussion before deciding it must move.
Explore the purpose-to-metric explanation →A new capability does not automatically justify a new destination. The distinction depends on the goal and organizational level you are examining. “We do not know yet” is a useful answer when paired with a next investigation.
Compare the responses and their risks ↓Examine a real strategic decision: Adobe →Enough for which commitment? A small, reversible experiment and an organization-wide change of direction should not carry the same burden of evidence. Match confidence to the scope, cost, obligations, and difficulty of reversing the decision.
Separate a capability demonstration from observed adoption, customer benefit, and viable economics. Record the source, date, sample, and limits. Look for evidence that could contradict the proposed direction, and compare similar situations where possible.
Before committing: write what you know, what remains assumed, the smallest useful test, and the result that would make you revise or stop. Set the review period long enough to observe the outcome. Early uptake cannot establish durable value by itself.
Record evidence and unknowns in a brief →Name whose outcome is being examined and who has authority over the relevant trade-offs. A team can surface evidence that challenges the wider direction without owning the decision to change it alone.
Identify the outcome owner, the teams and customers affected, the wider purpose the outcome supports, and the decisions this group can actually make. Bring dependencies and obligations into the discussion before changing what others are expected to deliver.
A useful opening: “This is the outcome we own. This is the wider outcome it supports. Here is what we can test, and here is who needs to join a decision beyond our scope.”
Define the scope and people involved →Leave with a provisional response and a learning commitment. Name the assumption to test, the outcome to observe, the trade-offs to protect, an owner, and a review point. Keep the test bounded by the authority and evidence you actually have.
At the review, compare what happened with what you expected. Did the customer or organizational outcome improve? What did it cost? What contradicted the assumption? Decide whether to continue, revise, stop, or reopen the goal.
A faster workflow, a downloaded brief, or a completed meeting is an activity. It does not establish that the decision helped. Return to the evidence and make the next commitment explicit.
Run the 25-minute team discussion →Prepare a brief and a review point →Try it: choose one outcome your team is rewarded for, then ask which assumption would have to stop holding for you to reconsider it. Practice with a fictional decision first →
CONNECTED NORTH STARS
Teams can pursue different guiding outcomes that contribute to a wider purpose. Make those connections explicit: every team can hit its target while the organization misses its purpose.
ILLUSTRATIVE EXAMPLE / CUSTOMER SERVICE
Customers resolve their needs with confidence.
Guiding outcome: Resolve problems without repeat contact.
Test the connection: Does faster handling help customers resolve their problems?
Guiding outcome: Enable reliable resolution across channels.
Test the connection: Do healthy systems still leave customers stranded between channels?
Guiding outcome: Automate help that produces durable resolution.
Test the connection: Are more conversations automated while unresolved needs increase?
Know whose decision it is. A team can surface evidence that challenges the wider direction without having authority to change it alone. Name who owns the outcome, which other teams depend on it, and who must join the decision.
Define the scope of your brief →A COMPARISON, NOT A SCORE
The response should match what changed. None is a higher level of maturity. “We do not know yet” can be the most honest starting point.
Stay
The guiding outcome and its underlying assumptions still hold. The constraint is execution, or the new signal has not yet demonstrated a meaningful shift.
An AI interface looks impressive, but a small customer test finds no improvement in task completion. Customers still need the same outcome and report reliability as their main problem.
Keep the reliability work on track. Reopen the decision if repeatable tests show materially better customer outcomes or different economics.
Watch for: Treating a decision to stay as permission to stop watching the evidence.
Adjust
Something meaningful has changed in how value can be delivered, but the outcome customers need still holds.
AI drafts support replies faster, but complex problems still require human escalation. Durable issue resolution remains the goal; the handoff and routing need redesign.
Compare repeat contact, resolution quality, and escalation outcomes. Faster replies alone are not evidence of better resolution.
Watch for: Preserving the goal in words while rewarding a proxy, such as automation rate, that pulls work away from it.
Shift
Evidence challenges whether the guiding outcome still represents the value that matters. Investigate a different definition of success before committing to it.
A team optimizes website conversion while customers increasingly complete purchases through other interfaces. The team must examine whether its channel-specific goal still captures customer and business value.
Test customer outcomes and economics across paths. Establish whether this is only a measurement gap or a deeper change in what the organization should pursue.
Watch for: Declaring a strategic pivot from a demo or a small pilot before understanding adoption, economics, and costs.
If your purpose is durable issue resolution, changing a dashboard from response speed to repeat-contact rate may improve measurement of the same outcome. If the organization changes which outcome governs investment and trade-offs, that is a deeper question.
Ask: “Are we measuring the same value better, delivering it differently, or reconsidering which value should guide us?”
THE METHOD
After acting, return to the evidence. Did your assumption hold? What should happen next?
Read the full argument →