Product strategy and prioritization

What should a strategic diagnosis establish?

The main constraint or opportunity supported by evidence, such as poor activation in a specific segment or an expensive distribution channel. A broad ambition like 'grow faster' does not identify the problem to solve.

Why can a product work for one segment and fail for another?

Needs, willingness to pay, workflow, procurement, support requirements, and alternatives differ. Evaluate value and delivery cost by segment rather than treating total market size as proof of fit.

What makes a useful initial market segment?

A coherent need, reachable users, a credible advantage, and a feasible delivery model. A narrow segment is helpful when it concentrates learning and value, not merely because it is small.

How can a bottom-up market estimate be constructed?

Estimate eligible accounts or users, realistic adoption, and value or revenue per account under explicit assumptions. Separate the reachable near-term market from a broad theoretical total and show sensitivity to the uncertain inputs.

Why include manual work and doing nothing in competitive analysis?

Users allocate effort and budget against those alternatives too. A product can outperform named vendors yet fail because the existing workaround is cheap, trusted, or adequate.

What makes a differentiator commercially meaningful?

Target users value it, can recognize it, and will change behavior because of it. A technically unique feature with no decision impact is not a demonstrated market advantage.

Why can a better product still fail to replace an incumbent?

Migration, training, integrations, procurement, trust, and coordination impose costs. The expected improvement must outweigh those costs, or the strategy must reduce them through onboarding and compatibility.

What should positioning clarify for the intended buyer?

The target context, relevant alternative, distinctive benefit, and evidence supporting it. Avoid claims so broad that users cannot tell when the product is appropriate or why it differs.

Why map the buyer's workflow beyond the feature being built?

Value may depend on upstream data, downstream approvals, or another team's adoption. Solving one screen can fail if a surrounding handoff remains the binding constraint.

Why should distribution be tested alongside product value?

Users must be reachable at a sustainable cost and through a credible channel. Strong satisfaction among a few recruited testers does not establish repeatable acquisition or sales.

What indicates that an acquisition channel may not scale?

Costs rise, reachable audiences saturate, lead quality falls, or conversion depends on unusually intensive manual work. Evaluate marginal performance as volume grows, not only early average results.

What should a product-level contribution calculation include?

Revenue attributable to usage or accounts minus relevant variable delivery, support, transaction, and servicing costs. State which fixed costs are excluded so positive contribution is not confused with company profitability.

Revenue per task is 10 and variable delivery cost is 4. What is contribution before other costs?

Six per task, or 60% of revenue. Include the costs relevant to the decision and do not assume that margin covers acquisition, fixed engineering, or overhead unless they were included.

Why segment cost to serve rather than using only an average?

Heavy users, complex integrations, and support-intensive customers can have very different economics. A profitable average can hide a segment whose growth increases losses or operational load.

Why can a lifetime-value estimate be fragile for a young product?

Retention, expansion, margins, and discounting are uncertain, and short cohorts do not reveal a full lifetime. Use scenarios and observed contribution rather than treating an extrapolated LTV as a settled fact.

What does acquisition payback measure?

How long cumulative contribution from an acquired cohort takes to recover its acquisition cost under a defined method. It depends on margin, retention, and revenue timing; revenue alone is not cash or contribution recovery.

What should a pricing unit align with?

Customer-understood value, predictable purchasing, and the product's cost/risk structure. Per-seat, per-use, and outcome-based pricing create different incentives; choose based on the workflow rather than copying a competitor.

What tradeoff comes with usage-based pricing?

It can align charges with consumption but make bills unpredictable or discourage valuable use. Provide understandable units, visibility, and controls, and test whether the unit tracks value rather than merely infrastructure cost.

How does packaging differ from choosing a price?

Packaging determines which capabilities, limits, service levels, and user rights are included together. It shapes who can adopt and upgrade; a price experiment cannot fully diagnose a poorly defined package.

What should justify a free tier?

A measurable role in acquisition, learning, network value, or conversion, with bounded cost and abuse exposure. Free usage is not automatically a growth engine if it rarely leads to sustainable value.

When can a large customer's custom request be strategically worthwhile?

When its value, reuse, learning, and commitment outweigh delivery and long-term maintenance costs. Document product divergence and opportunity cost rather than treating account size as automatic approval.

How should a roadmap distinguish commitment from exploration?

Label near-term deliverables with scope and dependencies, and later opportunities with confidence and learning milestones. Avoid presenting uncertain bets as dated promises merely to make the roadmap look precise.

What is a scoring framework useful for, and what is its limit?

It makes assumptions about value, reach, confidence, and effort explicit for comparison. Scores are not objective truth; inspect input uncertainty, strategic constraints, and whether small changes reverse the ranking.

What should be checked before comparing RICE scores across projects?

That reach uses the same period, impact scales mean the same thing, confidence reflects evidence, and effort includes relevant work. Inconsistent units can make the arithmetic precise but the ordering meaningless.

How does cost of delay differ from total expected value?

It describes what is lost by waiting, which may depend on a deadline, compounding effect, or temporary opportunity. A valuable project can tolerate delay while a smaller time-sensitive one cannot.

Why might the highest-value feature not be built first?

A prerequisite, migration, or learning step may determine whether it can succeed. Sequence work to reduce critical uncertainty and unblock value rather than sorting independent scores when the work is interdependent.

What is the value of a small reversible investment?

It can buy information or preserve a later choice without committing the full cost. Define what decision the option enables and when it expires; small projects are not valuable merely because they are cheap.

Why should past spending not be the main reason to continue a failing project?

The decision concerns future expected value versus future cost and alternatives. Past work may create reusable assets, but unrecoverable spending does not become a benefit of spending more.

What makes a project's stop criteria useful?

They specify evidence, time or resource limits, and who decides when the core assumptions fail. Write them before attachment grows, while allowing justified revisions when genuinely new information changes the decision.

Why balance a product portfolio across different uncertainty levels?

Reliable improvements, capability investments, and exploratory bets have different horizons and risk. The mix should reflect strategy and capacity rather than giving every team an identical percentage without context.

When should several teams invest in a shared capability?

When recurring common needs and adoption evidence justify the build, migration, and governance cost. Identify owners and consumers; speculative future reuse can create a platform nobody actually uses.

What belongs in a build-versus-buy comparison?

Capability fit, integration, operating effort, total cost, data controls, reliability, vendor dependence, and exit cost. Buying shifts work and risk rather than eliminating them, while building requires long-term ownership.

What makes an exit plan credible?

Document export formats, data ownership, replacement interfaces, migration steps, costs, and continuity requirements. Test a small export or substitution; contractual portability without usable data or tooling may be insufficient.

How does a network effect differ from ordinary scale economies?

A network effect changes value to users as participation changes; scale economies reduce average cost as output grows. They can coexist, but growth alone proves neither and can worsen congestion or quality.

Why can increasing supply fail to improve a marketplace?

Supply may be in the wrong locations, times, or quality bands, or demand may be insufficient. Measure successful matches and user outcomes by market rather than celebrating total listings.

How should a new tier that shifts users from an old tier be evaluated?

Measure incremental value and contribution across the portfolio, including retention and migration effects. Revenue attributed to the new tier may partly replace existing revenue rather than represent growth.

Why is dependence on one large customer a product-strategy risk?

Their churn, negotiation power, or custom requirements can dominate revenue and roadmap decisions. Evaluate concentration alongside contract quality, diversification cost, and the deliberate market strategy.

How can reliability work compete fairly with a new feature?

Estimate avoided user harm, support cost, retention risk, delivery friction, and contractual exposure. Use evidence and uncertainty; reliability need not be justified only by an immediate feature unlock.

Why is accessibility part of product strategy rather than only a final checklist?

It affects who can use the product, task success, purchasing requirements, and maintenance cost. Include relevant users and constraints early so adoption is not limited by avoidable design barriers.

What should a strategic refusal communicate?

Which users and outcomes the team is prioritizing, why the request does not currently fit, and what evidence could change that. Do not invent a future delivery date to soften an actual no.

Growth and support teams disagree about a launch. What should the decision compare?

Expected user value, support demand, quality risk, capacity, and reversibility under shared assumptions. Make the tradeoff and accountable decision explicit rather than resolving it through the loudest stakeholder.

Why does adding engineers not necessarily accelerate a late project?

Onboarding, coordination, scarce review capacity, and sequential dependencies can dominate. Identify the bottleneck and remaining work before treating headcount as interchangeable throughput.

What evidence should trigger a strategy review?

Material changes in user needs, distribution, economics, competition, constraints, or the assumptions behind the plan. Routine noise need not overturn strategy, but a calendar review should not delay responding to decisive evidence.

How do scenarios improve a product investment decision?

They show outcomes under plausible adoption, cost, retention, or delivery assumptions. Identify which variables drive the decision and what evidence would narrow them rather than presenting one optimistic forecast as certainty.

What separates a pilot from evidence of a repeatable business?

A pilot may rely on special pricing, custom support, or unusually motivated users. Test renewal, standard onboarding, repeatable acquisition, and cost to serve before generalizing from pilot satisfaction.

What should a feature-retirement plan include?

Affected users and workflows, migration or export options, notice, support, dependencies, and a final verification of removal. Low usage alone does not reveal the impact on a small critical group.

How should strategy be monitored without reducing it to one target?

Track outcomes, leading assumptions, economics, and important guardrails for the chosen market. Review whether the measures still represent value and investigate conflicting signals rather than optimizing one number blindly.

What makes a learning milestone different from a delivery milestone?

It resolves a decision-relevant uncertainty with specified evidence, such as whether buyers can complete procurement at the proposed terms. A prototype shipped without an answered question may meet delivery but not learning goals.

What should a strategy memo make reviewable?

The diagnosis, target users, chosen advantage, alternatives rejected, investment sequence, assumptions, and success or reconsideration criteria. A reviewer should be able to challenge the causal logic and resource commitments.