Plans · Software

Run a round of customer research

6 weeks · 4 phases · 23 tasks · 4 milestones · 4 risks

For a product team that wants to talk to customers properly before deciding what to build next. The weeks run from framing the questions and recruiting participants through interviews to synthesis and a readout that leads to decisions. Adapt the number of interviews to what you're trying to learn, and keep the questions tied to a decision you actually need to make.

Open this in Harbor

Makes it a program in your harbor, dated from today.

Milestones

  1. Week 1Research plan agreed
  2. Week 2Sessions booked
  3. Week 4Interviews done
  4. Week 6Readout delivered

The plan, phase by phase

1. Frame

Week 1. Decide what you need to learn and why.

  • Week 1Write the decisions this research should informIf no decision depends on the answer, the question can wait.
  • Week 1Turn decisions into research questions
  • Week 1Decide who to talk to and how manyFive to eight people per segment usually shows the main patterns for a question like this.
  • Week 1Agree the plan with stakeholders
  • Week 1Review existing research and support ticketsStart from what you already know so interviews go deeper.

2. Recruit

Week 2. Find the right participants and book them in.

  • Week 2Write the screener questions
  • Week 2Send invitations to customers and prospectsInvite more people than you need: some won't reply and some won't show.
  • Week 2Arrange incentivesGift cards or account credit, sized to their time. Check any rules your company has on gifts.
  • Week 2Prepare consent forms and recording permissions
  • Week 2Schedule sessions

3. Interview

Weeks 2 to 4. Run the sessions and capture what you hear.

  • Week 2Write the discussion guideAsk about past behavior, not opinions about the future.
  • Week 3Run a pilot interview and adjust the guide
  • Week 3Run the interviewsOne person asks, one takes notes. Leave silences: people fill them with the useful parts.
  • Week 3Invite engineers and designers to observe
  • Week 4Write notes the same day as each sessionQuotes and observations while they're fresh, kept separate from your interpretation.
  • Week 4Send thank-you notes and incentives

4. Synthesize and share

Weeks 5 to 6. Find patterns, agree what they mean and act on them.

  • Week 5Group observations into themesPut each observation on its own note and cluster them. Count how many people mentioned each theme.
  • Week 5Run a synthesis session with the team
  • Week 5Write findings with supporting quotesSeparate what you heard from what you think it means.
  • Week 6Present the readout and agree decisionsEnd the readout with the decisions made and who owns each one.
  • Week 6Store recordings and notes where the team can find them
  • Week 6Add follow-up work to the backlog
  • Week 6Share a short summary with participants who asked

What usually goes wrong

  • Recruiting the wrong people

    Talking to friendly power users or the wrong segment gives answers that don't generalize.

    What helps: Use a screener tied to the research questions and include people who left or never bought.

    Likelihood 3 of 5 · impact 4 of 5

  • Leading questions

    Asking whether people would use a feature invites polite yeses.

    What helps: Ask about what they did last time, pilot the guide, and have a second person review it.

    Likelihood 4 of 5 · impact 3 of 5

  • Findings don't lead to decisions

    A report nobody acts on wastes the participants' time and the team's.

    What helps: Tie each research question to a decision up front and end the readout by making those decisions.

    Likelihood 3 of 5 · impact 4 of 5

  • Recruiting takes longer than planned

    Busy customers reply slowly and no-shows leave gaps in the schedule.

    What helps: Start recruiting in week 1, over-recruit, and keep a short list of backups.

    Likelihood 4 of 5 · impact 2 of 5

Using it

Copy what fits and cut what doesn’t: no real project matches a template. In Harbor the plan becomes a program with these milestones, one project holding every task with its due date, and these risks in the register, all yours until you hand them out. The first phase starts in To do and the rest wait in the backlog.

Open this in Harbor

More plans

See every plan