A practical decision framework

  1. 1. Define what stalled means

    Replace “launch is stalled” with a concrete description: a release cannot be approved, users do not understand the offer, a key workflow is incomplete, or nobody owns the next decision. Clarify which launch outcome the team is pursuing and what is already complete. Different people may use “launch” to mean internal release, public announcement, sales readiness, or broad availability. Agree on the intended event and audience before diagnosing a delay. Otherwise, teams can report progress against different definitions.

  2. 2. Map the end-to-end launch path

    Trace the journey from intended audience to product discovery, evaluation, onboarding, first use, support, and follow-up. At each stage, note the owner, required input, current evidence, and handoff. Include product, marketing, sales, support, operations, and technical release steps only where they apply. The map should reveal where work waits or where the experience breaks, not create a checklist for its own sake. Distinguish a missing deliverable from a dependency that is not yet available.

  3. 3. Check product and release readiness

    Walk through the core user path using the actual release candidate and environment. Verify that required behavior, permissions, data, integrations, and error handling are understood. Check who can approve and execute a release, how the team will observe behavior, and what recovery option exists. A feature that works in a demonstration may still have unresolved operational or support needs. Record reproducible defects separately from preferences and future improvements so the launch decision has a clear basis.

  4. 4. Check audience, promise, and proof

    Ask whether the intended user can recognize the problem, understand what the product does, and see what action to take next. Compare public language with working product behavior and available evidence. Identify claims that need substantiation or wording that implies capabilities not present. If messaging is unclear, test comprehension with representative readers before expanding promotional activity. If the promise is accurate but the product cannot deliver it, the problem is readiness rather than copy.

  5. 5. Check ownership and decisions

    List launch decisions that are open, the person accountable for each, the information needed, and when the decision affects the schedule. Common gaps include unclear audience, unresolved scope, missing support ownership, approval uncertainty, and dependencies without an owner. Avoid treating every delay as an execution failure. A team may be waiting because the decision criteria are missing or stakeholders disagree. Make the unresolved choice explicit and gather only the information needed to decide.

  6. 6. Classify the constraint and respond

    Use the evidence to classify the active constraint as product, technical, message, operational, or decision-related. More than one may exist, but identify the one currently blocking the next step. Choose a focused action: repair a critical path, clarify a claim, assign support ownership, test a message, or resolve a decision. State the completion evidence and owner. Do not restart the entire launch plan when a smaller correction can clear the path.

  7. 7. Set a review and learning loop

    After the next step, review what changed and what remains uncertain. Capture questions from users and internal teams, route them to an owner, and distinguish launch feedback from defects or future requests. Update the launch path when evidence changes. Keep the plan connected to the product roadmap and support process so the team can respond after release. The point of diagnosis is a reasoned next action, not a claim that every risk can be removed before users encounter the product.

Keep a short decision log as the launch resumes. Record the blocker, evidence gathered, choice made, accountable owner, and any condition that would reopen the decision. This helps teams avoid cycling through the same debate and makes handoffs clearer when responsibilities shift. Revisit the log after the next release checkpoint or when a material assumption changes. Avoid turning the log into a second project plan; link to the working task system and preserve only the context needed to understand the decision.

Choose the next action

Use the table to make the current decision explicit. The categories are prompts for discussion; actual evidence and context should determine the path.

Signal Likely focus Next move
Core behavior is understood and constraint is bounded Focused intervention Assign an owner, define verification, and address the specific blocker.
Direction or user need remains uncertain Product learning Test the assumption that would change scope before committing to broad delivery.
Multiple paths depend on unclear foundations Technical or operating decision Map dependencies, compare options, and agree on sequencing and ownership.

For related work, see the Launch Strategy and How to Build a Startup Product Launch Plan. Use these pages as context; the framework here is general guidance and should be adapted to the team’s product, obligations, and constraints.

Questions teams ask

Is a stalled launch a marketing or product problem?

It may be either, or an operational or decision issue. Compare the promise with actual product behavior, then locate where the path breaks: unclear audience or message points toward marketing, while an incomplete core workflow points toward product readiness.

Should we delay a launch if some features are unfinished?

Decide against the intended user outcome and the risks of the available release. Identify whether unfinished work blocks a critical workflow, creates an unsupported promise, or can be clearly deferred without undermining use or support.

What should a launch blocker include before it is escalated?

Describe the blocked step, the evidence that it cannot proceed, its effect on the intended launch, and the decision or dependency needed to move forward. Name an owner and the next point at which the issue will be reviewed.