← Back to blog

Discovery Workshop Agenda: A Timed Plan You Can Run

November 13, 2026
Discovery Workshop Agenda: A Timed Plan You Can Run

A good discovery workshop agenda does one thing: it moves a group from "here is what we think we know" to "here is what we decided, who owns it and what happens next." The plan below does that in six blocks: frame the goal, surface what is known, map users and needs, test assumptions, prioritize, and commit to next steps. It comes with timings for a one-day session, a three-hour virtual version, the questions to ask and the follow-up that makes the day count. If you run these sessions as an independent project manager, The Freelance PM Club gives you a place to practice them with peers and coaches.

The Freelance PM Club
Build Your Freelance PM Practice
The Freelance PM Club offers expert-led courses, career coaching, curated job roles, and a community for independent project managers.
Visit

TL;DR

  • Start from the output, not the schedule: decide what artifacts and decisions must exist when people leave, then build the agenda backward.
  • Keep the room small (roughly 4 to 12 people, depending on the format) and cross-functional, with one facilitator who does not also argue a position.
  • A reliable arc is frame, share evidence, map users and needs, surface assumptions, prioritize, commit. Every block feeds the next.
  • Virtual sessions need fewer activities, a real warm-up, a break and a few minutes of slack for technology.
  • Send a one-page pre-read beforehand and a short summary with action items afterward.

Table of Contents

What a discovery workshop is for

Nielsen Norman Group describes the discovery workshop as a session that builds consensus on project milestones and plans, helping the team understand business requirements, collect existing knowledge and agree on priorities. It typically happens at the start of a project, or at the start of a major phase, when you need to absorb existing research, learn what stakeholders expect and build a shared view of direction.

For a freelance PM, that last point matters most. You are often the newest person in the room, and the workshop is the fastest way to hear the sponsor, the product owner, the designer and the engineer say their versions of the project out loud, in front of each other.

One caution on scope. A workshop is a facilitated working session where a team makes sense of what it knows about customers, frames opportunities and agrees how to learn next. Learning Loop points out it does not replace customer interviews or tests. Use a workshop when several people need a shared understanding to make a decision. If everyone already agrees on the problem and a quick customer call would settle the open question, make the call.

Set the goal, outputs and attendees first

The agenda is the last thing you write, not the first. The International Labour Organization advises building the agenda only after you know the primary objectives and who will attend, and it wants objectives phrased as observable behaviors or actions, so you can tell afterward whether they were met.

Write the purpose in three lines

Before any scheduling, answer three questions in writing:

  1. Why is this workshop happening now?
  2. What should be different when people leave?
  3. What happens if we skip it?

If you cannot answer the second one, you are not ready. Toby Sinclair's planning template puts it bluntly: if you cannot say what should be different when people leave, do not book the room.

Define the outputs

A workshop without a product is a meeting. List the artifacts you expect to leave with. Typical discovery deliverables, according to DO OK, include a value-driven pyramid, lean canvas business models, a high-level MVP scope, user personas and journeys, and a project plan with estimates. Iterators adds SMART goals, a value proposition, a timeline and budget estimate, and success metrics.

You will not produce all of that in one day. Pick three. For a typical freelance engagement, a sensible three are: a one-sentence problem statement, a ranked list of risks and assumptions, and a draft scope with owners.

Choose who is in the room

Guidance on headcount varies by format. DECODE recommends 4 to 7 participants built around a product manager, UX/UI designer and solution architect. WAITRO suggests 6 to 12 across stakeholder groups, areas of expertise and levels, and warns that too many senior executives can inhibit open discussion. The same source argues that workshops often fail because the mix is too similar, which creates an echo chamber that reinforces existing assumptions.

A practical rule: one decision-maker, one person who knows the customer, one who knows the technology, one who knows the money, and you as facilitator. Add voices until the group stops being able to hear itself think.

Choose who is in the room

The pre-read: what to send before the day

Learning Loop recommends a one-page pre-read sent one or two days before the session. It covers the outcome you want, the decision to be made, existing evidence, open questions, the agenda, participant roles and links to raw sources. Prepare a shared board with columns for Outcome, Observed evidence, Customer opportunities, Possible solutions, Assumptions and Next research.

The pre-read does two jobs. It saves the first hour from becoming a briefing, and it lets quieter people prepare their thinking. Our guide to workshop facilitation for independent PMs covers the logistics around it.

A one-day discovery workshop agenda

This schedule is built for six to eight people and roughly six hours of working time. It follows the common shape of published templates: airfocus's template, for example, moves from vision to KPIs, personas, user stories and story mapping with breaks between blocks, and Google's facilitator guide for a two-day remote format starts by understanding the space, then finds opportunities, then designs solutions. The timings below are a starting point to adapt.

TimeBlockWhat happensOutput
9:00 to 9:20Welcome and ground rulesPurpose, outputs, roles, how decisions get madeShared purpose
9:20 to 10:05Vision and contextSponsor presents the vision; team questions itProblem statement draft
10:05 to 10:35Evidence reviewWhat research, data and past attempts exist?Evidence column filled
10:35 to 10:50Break
10:50 to 11:35Users and needsPersonas, jobs to be done, current journeyPersona and journey sketch
11:35 to 12:15Business and technical constraintsBudget, timeline, systems, regulationsConstraint list
12:15 to 1:00Lunch
1:00 to 1:45AssumptionsSilent writing, then cluster and rate by riskRanked assumptions
1:45 to 2:30Opportunities and prioritizationFrame opportunity statements; voteTop 2 to 3 priorities
2:30 to 2:45Break
2:45 to 3:30Scope and success measuresDraft scope, KPIs, ownersDraft scope
3:30 to 4:15Next stepsLearning actions, risks, who does what by whenAction list
4:15 to 4:30Check-outOne-line reflections, feedbackFeedback

Why this order works

The sequence follows the three areas a discovery effort has to cover: business, users and technology. Surfacing them separately keeps the loudest function from setting the frame. GitLab's handbook gives the test for every block: the output of one block feeds the next, and if you cannot say in one line why an activity is there, cut it.

A shorter agenda: 90 minutes

When a day is not available, Learning Loop's five-step flow fits in roughly 90 minutes: frame the outcome and review evidence (25 minutes), collect assumptions (15), map customer opportunities (20), prioritize the problem to focus on (15), and plan the next learning action (15). That adds up to 90.

Five timed discovery steps fill a 90-minute session

This suits a mid-project check or a small engagement where the sponsor will give you one morning, not a week. For a kickoff rather than discovery, see our 60-minute project kickoff agenda.

Exercises that surface business, user and technical needs

Pick one or two per block, not all of them.

  • Silent brainstorm. Everyone writes on sticky notes or board cards for five to ten minutes before anyone speaks. It stops anchoring and gives introverts equal weight.
  • Opportunity statements. Turn observations into "users struggle to ___ because ___." Google's facilitator guide uses this step before choosing an opportunity to pursue.
  • Assumption mapping. Plot assumptions on two axes: how critical to the plan, and how little evidence you have. The top-right quadrant becomes your research plan.
  • Journey sketch. Draw the current process end to end, marking where the customer waits, repeats work or drops out.
  • Dot voting. Give each person a fixed number of dots. It is fast, visible and good enough for a first ranking.
  • Split the pairs. Researchers writing on running interactive workshops note that splitting colleagues who usually work together exposes them to other perspectives and stimulates new thinking. They also advise photographing every artifact on the day, since paper is easily lost.

If risk is the main worry on your project, run the dedicated risk workshop as a follow-on session rather than cramming it into discovery.

Questions to ask during a discovery session

Ben Aston's list of 24 discovery questions groups them into six types: strategy, user experience, content, creative, technical and project management. He is clear that it is a framework for understanding the client's problem, not a questionnaire to hand over. Airfocus lists the core ones: what is the purpose of the product, what problem are you solving, how will you differentiate, who are your users and what matters to them.

A compact set for a workshop:

Strategy

  • What problem are we solving, and for whom?
  • What does success look like in six months?
  • What happens if we do nothing?

Users

  • Who uses this today, and what do they do instead?
  • What have customers told us, and how do we know?

Technology and constraints

  • What systems must this work with?
  • What is fixed: budget, date, vendor, regulation?

Project

  • Who signs off, and who can veto?
  • Who has worked on this before, and what did they learn?

Use follow-ups like "Can you give me a recent example?" or "What did that cost?" when an answer matters. If your session is a one-to-one conversation with a prospective client rather than a group, the live-call version of these questions is in our guide to discovery call questions.

Adapting the agenda for a virtual workshop

Virtual works, but not at the same density. Choice for Youth's facilitation guide warns that packing an online agenda to match an in-person day leads to rushed conversations and shallow engagement. It recommends a proper 30-minute warm-up, a mid-session break, a quick energizer afterward and 5 to 10 minutes of slack for technical problems.

GitLab's handbook suggests 60 to 90 minute cross-functional workshops on video with a shared board, breakout groups of two or three, and rough proportions of 15 to 20 percent for framing, 20 to 25 minutes per diverge block, 10 to 15 minutes to converge and share, and 10 to 15 percent to close.

Here is a three-hour virtual agenda adapted from Workshop Weaver's timed plan:

TimeBlock
0:00 to 0:10Welcome, tech check, ground rules
0:10 to 0:20Check-in round
0:20 to 0:30Context presentation
0:30 to 0:50Silent brainstorm on the board
0:50 to 1:10Breakouts: cluster and discuss themes
1:10 to 1:25Share back top three themes
1:25 to 1:40Break
1:40 to 1:55Dot voting
1:55 to 2:15Breakouts: action planning
2:15 to 2:35Share back and discuss
2:35 to 2:50Decision round
2:50 to 3:00Check-out and feedback

For teams across time zones, Mural's remote facilitation guide advises listing start times in each person's local time zone and using durations in the agenda, and splitting work between two facilitators when possible: one runs the room, the other watches chat and tools. If you are weighing how much of the work should happen outside live time, see our piece on asynchronous project management and the remote kickoff templates built to cut video fatigue.

Close the loop: follow-up that makes the day count

Within 24 hours, send a concise summary written for someone who was not there. GitLab's guidance is to pull out the top moments or concepts and open questions, then create and share action items and artifacts. Include:

  1. The problem statement as agreed.
  2. The ranked priorities and what was deliberately left out.
  3. Open questions with an owner and a date.
  4. Photos or exports of every board.
  5. The next touchpoint.

Then move straight into onboarding the engagement. Our client onboarding checklist picks up where the workshop ends, and the guide to client expectation management helps you hold the scope the room agreed on.

A reusable discovery workshop template

Copy this into your own doc and fill it in per client.

  • Purpose: Why now? What changes if we succeed?
  • Outputs (pick three): problem statement, ranked assumptions, draft scope, personas, KPI set.
  • Attendees and roles: facilitator, scribe, decision-maker, customer voice, technical voice.
  • Pre-read sent: date, link, raw sources attached.
  • Agenda blocks: time, activity, owner, output.
  • Parking lot: where off-topic items go.
  • Follow-up: summary sent by, action owners, next session.

Perspective

Most independent PMs do not fail in discovery because they lack a template. They fail because they run their first workshops alone, with no one to rehearse against and no one to tell them the agenda was too full.

That is why we built the Club around the whole pathway: curated roles to find the work, courses to build the skills, coaching to sharpen how you run the room, and a community of project managers who have faced the same sponsor who wants the answer before the question. Our members are freelancers, consultants and people leaving corporate roles, and a workshop is often the first moment they have to show a client what they are worth.

Our view: practice the agenda, borrow the templates, get feedback from people in the same seat, and walk into your next discovery session with a plan you have already tested out loud.

Ashanti

Where The Freelance PM Club fits

A workshop agenda is the visible part of the job. The harder part is finding the engagements, building the facilitation skill and having people to compare notes with. The Club combines a job board, expert-led courses, a template and tool library, a networking community and coaching for freelance project managers and consultants. Membership plans are listed at $5 per month for Job Board Access, $10 per month for Freelancer Pro and $20 per month for Freelancer Elite, with details on the subscriptions page.

The Freelance PM Club

If you want a second pair of eyes on your first agenda, our coaching is built for independent PMs working through exactly these moments.

FAQ

How long should a discovery workshop be?

It depends on scope. A focused session can run 90 minutes using a five-step flow, a virtual workshop commonly fits in three hours, and a fuller in-person discovery runs about a day. Choose the shortest format that produces your agreed outputs, and add slack rather than activities.

Who should attend a discovery workshop?

A small cross-functional group: a decision-maker, someone who knows the customer, someone who knows the technology and a facilitator. Published guidance ranges from about 4 to 12 participants depending on format. Avoid filling the room with senior executives, since their presence can inhibit open discussion.

What is the difference between a discovery workshop and a kickoff meeting?

Discovery decides what the problem, priorities and scope are. A kickoff starts delivery once those are agreed, covering roles, communication and the first milestones. Many projects need both, with discovery coming first.

What should I do if the group cannot agree on priorities?

Use dot voting to make disagreement visible, then ask each person to explain their top choice in one minute. If two options stay tied, turn the tie into a learning action with an owner and a date, so the decision rests on evidence rather than rank.

Sources