# Exploration workshop A facilitated session to identify the problem, constraints and next decisions. **TL;DR** Find out what is actually wrong, what matters, and who can decide what happens next. Leave with a shared problem and a few useful next steps. The solution can wait until we know the question. Status: Workshop method ## Before you start For: Facilitators and project leads Use when: You need to understand an unfamiliar organization before proposing work. Bring: A decision-maker, permission to take notes and existing business or product material. Leave with: An evidence-backed account of the problem and a list of missing information. ## Worked example (fictional) A fictional lab asks for a dashboard. Exploration reveals that experts cannot agree on review criteria. The next step becomes a criteria workshop, not dashboard design. ## Check the result Separate the initial request from the underlying problem, and assign an owner to every missing input. ## Limitations One hour is an intake, not validated research. Ask only for material relevant to the decision; permission to meet is not permission to access every system. The first conversation should change your understanding of the work. Someone may arrive asking for a new website and leave with a much better question about what the website needs to do. Use this script to get around the company in an hour. Follow the answers that matter, skip what you already know, and write down what you still need to find out. The questions are there to help you listen. You do not have to finish the list. ## Prepare From their side: the founders or whoever decides, plus anyone they want in the room. From ours: whoever will run the engagement. Keep it small enough that people answer instead of presenting. Record the meeting and get it transcribed. The transcript is what the proposal gets built from. Ask permission to record at the start. One hour. Video call or in person, whichever happens sooner. Before the call, ask them to have documentation ready to share: deck, brand files, site analytics, anything they have. The goal is coverage, not depth: touch every area, note where it hurts, and go deep later in the workshop. Let them talk; the questions are prompts, not a form to fill. ## The company · ±10 min Start wide. The answers here decide which of the later blocks deserve the most time, and they produce the sentences the proposal quotes back. - What does the company do, in your own words, as you'd say it to a customer? - How do you make money, and who pays? - What stage are you at: funding, team size, runway? - What's working right now? What's clearly not? - What does success look like twelve months from now? - Why are you talking to us, and why now? - Who decides on this engagement, and who else needs to be convinced? ## Brand · ±10 min Find out what exists, whether it still fits, and how far the appetite for change goes. - How do you describe yourselves today, and does that still feel true? - Does the current identity feel like you? Where does it pinch? - What exists: logo files, guidelines, fonts, a design system? Who made them, when? - What do customers say you are? Does that match what you want to be? - Which brands do you admire, inside or outside your market? - Are you attached to the name? The logo? Anything untouchable? - Where does the brand have to live: site, product, deck, print, merch? ## Product · ±10 min Understand what users hire the product for and where it struggles. Even when the engagement is 'just' a website, the product decides what the story can honestly claim. - What do users actually hire the product for? - Walk me through a first use, from landing to first value. Where do people drop off? - What's on the roadmap for the next two quarters, and why those things? - Who builds it: team, stack, what's easy and what's expensive to change? - What would you cut if you had to cut something? - What do users ask for that you don't want to build? ## Marketing · ±10 min Map how customers arrive today and what the company claims, and whether those claims survive contact with proof. - Where do customers come from today, channel by channel, roughly what share? - What's the best-performing thing you've ever done? The most expensive failure? - Who is the buyer, and is that the same person as the user? - What claims do you make? Which can you prove with a name or a number? - Who do you lose deals to, and what reason do you hear? - What does marketing cost today: people, tools, spend? ## Ops · ±10 min How work actually moves through the company: where the hours go, what's manual, and what breaks under growth. - Walk me through how work moves from sold to delivered to invoiced. - What's manual that shouldn't be? What gets copied between tools by hand? - What depends on one person? What happens when they're out? - Which tools do you run on, and which do you pay for but barely use? - What breaks first if you double in size? - Where does AI already help you, and where do you suspect it could? ## The ask & documentation · ±10 min Close with logistics: what they can share, what happens next, and the boundaries that shape the proposal. Ask for everything. More documentation means better deep research behind the proposal. - The deck, and any past brand or strategy documents - Brand assets: logos, fonts, guidelines, design files - Site and product analytics access, or exports - Customer quotes, testimonials, won/lost deal notes - Any past research, audits, or agency work - Budget range and timeline expectations, stated out loud - A date: when they'll have the proposal, and who reviews it ## Output A transcript covering every area we could help with, the documentation set, and a stated budget and timeline. The transcript becomes the proposal (including the deep research on positioning, gaps, competitors, and visual quality), generally within days of the call. ## Handoff Record the input sources, decisions, assumptions, owner and acceptance checks. Have the responsible person approve the record before downstream automation. ## Tools, in order ### Before the session [Notion](https://www.notion.com/) · Paper or a shared workshop board Collect the initial request, existing work, and the people who can explain it. Put the decision you need on the first page. Agree on recording, transcription, access, and retention with everyone before starting; handwritten notes are a valid alternative. Carry forward: A brief, a prepared board, and an agreed capture method. ### In the room [Notion AI Meeting Notes](https://www.notion.com/help/ai-meeting-notes) · Paper or a shared workshop board · Phone camera Use the recorder to support listening, not to replace a facilitator. Ask for real examples of how work happens before proposing a solution. Write decisions and disagreements visibly; photograph the boards with permission. Pause recording whenever someone asks. Carry forward: Authorized source notes, a transcript if agreed, and labeled board photos. ### After the session [Claude](https://claude.ai/), [ChatGPT](https://chatgpt.com/), or [Grok](https://grok.com/) · [Notion](https://www.notion.com/) Check speaker names, quotations, numbers, and apparent agreements against the sources. Ask an assistant to draft a problem brief with evidence, constraints, and missing inputs, with a source ID for each decision and a separate list of open questions. Have participants correct the record before using it as a brief. Carry forward: a problem brief with evidence, constraints, and missing inputs ### Review the work with partners We make and revise the work with Claude, ChatGPT, or Grok, then put selected versions on our Studio project pages for partners to review. Check the preview before sharing it. You can do the same with a private prototype or shared document; use a workspace that can access the files you need. Name the version, say what changed, and ask the question you need answered. Keep feedback with that version and discuss conflicting requests before making the next changes. Confirm approval separately, and keep confidential work in a restricted space. For this method: Does this describe the problem you need solved, or only the solution you first asked for? ### Delegate the routine work Turn verified notes into a source-linked table, label board photos, or format the agreed action list. Give the agent only the relevant excerpts and a completed example. Keep with a person: Interpreting disagreement, choosing a direction, and deciding what participants actually approved stay with the facilitator and decision owner. [Sequence prompts, then delegate](https://noord.dev/polder/delegating-with-prompts) [Choose a model for the design task](https://noord.dev/polder/choosing-design-models) Source: https://noord.dev/polder/workshop-exploration Public working material from Noord. Third-party code, fonts and assets retain their own licenses. Publication does not imply a license to redistribute third-party material.