Most growth meetings begin with a channel. Stronger ones begin with a constraint.
The difference matters. A media team sees an acquisition problem. A CRO specialist sees a conversion problem. CRM sees retention. Merchandising sees product mix. Each view can be accurate and still compete for resources the business does not have.
A constraint is the limiting condition that most restricts valuable growth now. It may sit in demand, conversion, repeat purchase, margin, inventory, cash, or the team’s ability to execute. This session is a disciplined way to choose one without pretending the rest are unimportant.
What you’ll learn
- How to map the commercial system before choosing a marketing project.
- How to separate evidence from functional opinion.
- How to rank candidate constraints by impact, confidence, control, and time to learning.
- How to shape one focused pilot with a baseline and stop rule.
Prepare one page of evidence.
The session works best when the team arrives with a compact view of the business. Do not spend the first half reconciling dashboards.
Bring recent movement and a relevant comparison for:
- qualified traffic or reach;
- site conversion and the main funnel steps;
- first-order value and contribution;
- new versus returning customer mix;
- repeat behavior by a sensible cohort window;
- hero-product stock and fulfilment capacity;
- cash, service, or team boundaries that can limit action.
Use your own definitions. If a metric is incomplete, label it. If two systems disagree, show the disagreement. Uncertainty is an input to prioritization, not something to hide before the meeting.
Run the 45-minute constraint session.
Minutes 0–8: state the commercial goal and boundaries.
Name the outcome and period. “Grow” is too broad. “Increase first-order contribution over the next quarter without exceeding the current inventory plan” provides a useful decision frame.
Add the constraints the team must respect: cash exposure, minimum contribution, stock cover, service load, claim boundaries, budget, or founder capacity. These are not footnotes. They decide whether an opportunity is usable.
Minutes 8–18: map where value is being lost.
Move through the journey in order: reach, qualified visit, product consideration, checkout, contribution, repeat, and capacity. At each stage, ask two questions:
- What changed enough to matter?
- What evidence says this stage is limiting the goal?
Write observations as facts where possible. “Mobile checkout completion fell after the release” is an observation. “The site team broke conversion” is an accusation disguised as analysis.
Minutes 18–28: nominate up to three candidate constraints.
A candidate should connect an observed problem to a commercial consequence. “Creative” is too broad. “The hero concept is losing qualified response in the prospecting audience, raising the cost of new-customer demand” is testable.
Limit the list. If every problem becomes a candidate, the team has avoided prioritization.
Minutes 28–38: score the candidates.
Use a simple one-to-five score on four dimensions:
- Impact: if improved, how much can this move the stated goal?
- Evidence: how strongly do current observations support the diagnosis?
- Control: can the team change it without waiting on a major dependency?
- Time to learning: can a useful signal arrive inside the decision period?
The score is a conversation tool, not mathematics. A high-impact problem with weak evidence may deserve a diagnostic action. A well-evidenced problem outside the team’s control may require escalation rather than a marketing test.
Minutes 38–45: choose the constraint and write the first action.
Select one primary constraint. Record why it outranks the alternatives and what new evidence would change the decision. Then define the smallest action that can create commercial value or materially improve confidence.
Goal: the commercial outcome and period.
Constraint: the limiting condition.
Evidence: the observations supporting it.
Unknown: what could overturn the diagnosis.
Action: the smallest useful intervention.
Guardrail: the risk boundary.
Review date: when the decision returns.
A worked example: traffic was the tempting answer.
Imagine a food brand wants more D2C contribution. Paid media has room to spend, so the first proposal is to increase acquisition.
The map shows:
- Traffic is stable, and prospecting click costs have not materially worsened.
- Site conversion is healthy for the category’s own recent history.
- Most first orders use a low-margin starter offer.
- Few customers move from the starter product into the core replenishment range.
- Inventory can support more core-product demand.
More traffic would create orders, but it would pour customers into a weak transition. The primary constraint is the path from trial to the first valuable repeat purchase.
The first action could be a cohort-based diagnosis: compare product sequence, timing, service signals, and message exposure for customers who do and do not make the second purchase. The team then tests one intervention against a clear baseline.
This does not prove acquisition should stop. It says the next unit of attention is likely more valuable in the repeat path.
That is progress. Re-run the map when the evidence changes rather than turning one diagnosis into a permanent strategy.
Turn the constraint into a focused pilot.
A good pilot is smaller than a transformation and more complete than a task list. It needs:
- One baseline. State the current metric, period, source, and definition.
- One primary outcome. Choose the commercial movement that would support the diagnosis.
- Leading signals. Add the earlier behaviors that show whether the mechanism is working.
- Decision rights. Name who can recommend, prepare, approve, and execute.
- Guardrails. Set budget, margin, stock, brand, and operational limits.
- A review point. Decide in advance when to scale, adjust, stop, or gather more evidence.
A pilot does not need to solve the entire constraint. It needs to produce a valuable result or a clearer decision quickly enough to justify the next step.
Prioritization is the discipline of making one problem easier to solve without forgetting the rest of the business.
Frequently asked questions
What if the team cannot agree on the primary constraint?
Compare the evidence and name the disagreement. If two explanations remain plausible, choose the smallest diagnostic that can separate them rather than forcing consensus through seniority.
Should we always choose the highest-scoring candidate?
No. The scores expose tradeoffs; they do not replace judgment. A cash or brand risk may override a simple total, and weak data may make a diagnostic the responsible first move.
How often should we rerun the session?
After a meaningful result, material market change, stock shift, or strategic reset. Many teams can use a lighter check weekly and a fuller review monthly or quarterly.
Can the main constraint sit outside marketing?
Yes. Inventory, fulfilment, product fit, cash, service capacity, and decision speed can limit growth. Marketing should not amplify a problem the business cannot absorb.
Find the point that limits valuable growth now. Give it one owner, one test, and a date when the evidence gets another vote.