Startups thrive on speed. Small teams, tight timelines and limited resources make traditional, heavyweight design processes risky. For early stage ventures, the aim is rapid validation: learning quickly whether you are building something users actually want.
- What problem do we believe users have?
- Why is solving it valuable to them and to our business?
- What evidence would confirm or contradict this belief?
This lean product design approach prioritises fast learning and collaboration over lengthy design cycles, without discarding UX discipline.
In this article, I explain how I adapt design workflows for small teams, how this connects to lean startup thinking, and how a structured but lightweight 5-step process helps ensure we spend less time guessing and more time building.
Iteration as an ongoing process
Iteration is not a formal step that happens once. It is a behaviour that shows up throughout the process.
- Early iteration challenges assumptions and scope
- Mid stage iteration improves flow and comprehension
- Late iteration focuses on polish, resilience, and edge cases
Most iterations are small and fast. If a change takes more than a day or two, it usually means the problem was not framed clearly enough earlier.
The goal is not to keep iterating forever. The goal is to iterate just enough to be confident, then ship.
Lean UX principles in a startup context
Designing with lean startup does not mean discarding UX practice. Instead, it means adapting UX methods for speed and relevance.
As Laura Klein reminds us in UX for Lean Startups, “You are not your user, and your opinion is not more valid than theirs.” It is a reminder to ground decisions in evidence rather than internal assumptions, even when time is tight.
In practice, this means:
- Preferring lightweight design artefacts over heavy documentation
- Running continuous, small research loops rather than long formal studies
- Working closely with engineering and product in short, iterative cycles
- Being ruthless about prioritisation and scope
These principles underpin the 5-step process below.
The 5-step lean UX process
Each step has a clear goal, a realistic time frame, and a concrete output, allowing teams to learn quickly without sacrificing quality.
1. Understand
2 to 4 hours over 1 to 2 days
Goal
Confirm the problem is real and worth solving. Gather enough context to move forward with confidence. This work is rarely neat. It happens between meetings, async reviews, and short conversations. That is normal in startups. The priority is speed, backed by evidence.
Tasks
- Scan recent support tickets, app reviews, and customer feedback
- Look for recurring patterns, not edge cases
- Check analytics for obvious drop offs or broken steps
- Review two or three competitors to understand common approaches
- Stop when insights start repeating
Output
- 5 to 10 clear insights written in plain language, captured in Notion or FigJam.
NN/g resources
2. Ideate
Goal
Decide what success looks like for the user and agree on a single happy path. This step creates alignment across design, product, and engineering. Keeping it focused avoids weeks of rework later.
Tasks
- Write one or two simple user stories
- Map the ideal step by step flow in FigJam
- Keep one action per step
- List assumptions explicitly
- Define success metrics such as completion rate or time on task
- Walk the flow through with a developer or product owner
Output
- A single happy path flow with a short written summary.
- A single user flow diagram with a short written summary.
NN/g resources
3. Design and prototype
2 to 4 days
Goal
Define structure and interaction without getting distracted by visual polish. Design work often expands slightly as feasibility questions come up.
Tasks
- Create mid fidelity, mobile first wireframes
- Explore alternatives where decisions feel risky
- Use platform native patterns and conventions
- Establish hierarchy and accessibility early
- Keep everything tied back to the happy path
Output
- Annotated wireframes in Figma that clearly express the intended flow.
- Clickable prototypes for testing.
NN/g resources
- Wireflows: A UX deliverable for workflows and apps
- First principles of interaction design
- UX prototypes: low fidelity vs. high fidelity
- Use good prototype specifications to empower team collaboration
4. Test and iterate
2 to 4 days of calendar time
Goal
Testing and iteration are tightly coupled. We are not looking for perfection, we are looking for the biggest risks and the fastest fixes.
The actual testing is usually quick. Scheduling users, reviewing recordings, and agreeing on what to fix is what takes time.
Tasks
- Build a clickable prototype in Figma
- Test with 3 to 5 users, or run a structured internal review
- Focus on task success and understanding
- Capture issues directly in the design file
- Prioritise issues by impact on outcomes
- Fix usability blockers first and recheck critical changes
Output
- A prioritised list of issues and an updated prototype.
- An interactive prototype ready for validation.
NN/g resources
5. Deliver and learn
2–4 days (calendar time)
Goal
Support build quality and keep learning after release. This is not a single phase. It runs alongside development and continues post launch.
Tasks
- Support engineers with async feedback in Figma or Loom
- Run design QA on real devices
- Check responsiveness, errors, loading, and accessibility
- Treat edge cases and variants as separate mini cycles
Output
- A signed off build and a short list of follow up improvements or experiments.
- Annotated notes in Figma and a prioritised list of issues.
NN/g resources
No Comments.