The Team Watch Party
Whole team watches interviews live, predicts first and debriefs after, replacing the research report with instant alignment
- Difficulty
- Moderate
- Time to result
- ~days to results
- Steps
- 5
- Confidence
- 90%
A way to turn customer research into a team sport: the whole core team live-streams the interviews, takes manual notes, debriefs between sessions, and captures collective takeaways — eliminating the written report and producing shared conviction. Bookended by a pre-interview prediction round and an individual big-takeaways form, it defeats hindsight bias and generates momentum because everyone saw the same evidence together.
Origin
Originated in Margolis' early Gmail-team research at Google, which had invested in video conferencing, streaming and one-way mirrors so teams could watch. Carried into GV and refined across 300+ sprints with design partners Kate Aronowitz and Vanessa Cho facilitating the back room.
Core principles
- 01If the whole team watches, you never need to write or 'sell' a research report — alignment is automatic
- 02Manual note-taking (not AI) forces people to lean in and stay engaged
- 03Predicting outcomes beforehand creates a snapshot that later reveals how much was actually learned
- 04Debriefing between interviews captures patterns while they're fresh
- 05The core product team must attend everything; they own and build the outcome
How to run it
- 1
Collect predictions before interviews
Have each team member write down specifically what they think will happen and what customers will prefer, pushing past vague answers to concrete hypotheses.
Pro tip Force specificity ('customers will prefer ASAP delivery') — not 'we'll find out what they think' — so the later comparison is meaningful.
- 2
Live-stream one-on-one interviews to the team
The interviewer runs each interview solo on Zoom while the team watches a separate live stream, so participants face only the interviewer, never a wall of observer faces.
Pro tip Keep the participant in a genuine 1:1 — don't put 20 heads on their screen.
- 3
Assign rotating note-taking and a backchannel
Team members take turns manually noting in a shared doc, run a Slack backchannel for chatter, and a facilitator (design partner) passes the interviewer clarifying questions judiciously.
Pro tip Manually taking notes — no AI — keeps people from leaning out into email or Slack.
Watch out Be judicious passing questions to the interviewer; a question can send the interview in an unpredictable direction.
- 4
Debrief between each interview into a structured sheet
After each session, fill a debrief spreadsheet whose rows are the key things you set out to learn and whose columns are participants, led by the decider/product owner.
Pro tip These rows are the learning goals, not the interview questions — they're different.
- 5
Capture individual big takeaways, then compare to predictions
At day's end each person independently fills a big-takeaways form (top takeaways, revised bullseye definition, next steps), then the team reviews patterns and compares results to the opening predictions.
Pro tip The prediction-vs-outcome comparison beats hindsight bias, which makes teams wrongly feel 'we already knew that'.
Watch out Watch for the curse of knowledge: expert teams systematically overestimate how much customers know, how big a problem they perceive, and how ready they are to buy.
In the wild
Margolis notes his record watch party had 40 attendees and that he hasn't written a research report in about ten years, because the team captures the big takeaways collectively through a Google form and shared spreadsheet.
→ Instant team alignment and momentum ('let's go do the next thing') without the overhead or persuasion of a report.
Across studies, teams that insist they already talk to customers constantly go through the watch party and repeatedly conclude the results look very different from what they assumed and that they need more research.
→ Reveals blind spots (curse of knowledge) even in teams confident they know their customers.
Common mistakes
Using AI or a single note-taker so the team leans out
If people know notes are handled they disengage — checking Slack or email — and lose the shared conviction that makes the watch party work; manual, rotating notes keep everyone in it.
Skipping the prediction round
Without a pre-interview snapshot of expectations, hindsight bias makes the team feel the findings were obvious, hiding how much was actually learned and undervaluing the research.
Is it for you?
Best for
Distributed product teams that need shared conviction and alignment from customer research, not just a report to argue over
Not ideal for
Solo founders or situations where you can't get the core decision-makers to commit a full day of attention
From the transcript
“the make it a team sport is just is been this very powerful um tool for building consensus building alignment”
“I don't write a report I haven't written a report in I don't know 10 years”
“we do this thing before the watch party where we get everybody to predict what they think they're going to learn”
“there's this concept of the curse of knowledge”
From the episode
Identify your bullseye customer in one day
Michael Margolis (UX Research Partner at Google Ventures)