Briefstarter

UX design briefs for portfolio practice

Three complete UX briefs you can start today, written the way a client would scope them. Every client, product and situation on this page is invented: these are practice assignments for building a portfolio, not real commissioned work, and any case study you make from them should say so.

What a useful UX brief contains

A brief that is worth working from answers seven things before you open a design tool. If one is missing, decide it yourself and write your decision down — that choice is part of the work.

Client context
Who the organisation is, how it makes money and what it has already tried.
Problem
The thing going wrong today, stated as a situation rather than a feature request.
Audience
Who uses it, where, on what device, and what they already know before they arrive.
Constraints
Technical, legal, brand and accessibility limits you are not allowed to design around.
Deliverables
The specific artefacts expected: flows, screens, states, components, rationale.
Timeline
How long you have, which sets how deep the research can reasonably go.
Evaluation criteria
How anyone decides the result worked, written so it can be checked rather than argued.

Three UX briefs, easy to advanced

EasyApp UX / UI4 days

Reserving a book at a city library

Halden City Libraries, a six-branch public library service

Problem

Members can reserve books online, but the reservation lives three taps deep inside a catalogue built for librarians. Front-desk staff say most people give up and ask at the counter, then never learn the feature exists.

Target user

Members aged 16 to 70 on a phone, usually standing in a shop or on a bus, who know the title they want and nothing about library terminology.

Scope

Redesign the search-to-reservation path only: finding a title, choosing a collection branch, confirming, and seeing where the reservation stands. Account settings, fines and events are out of scope.

Constraints

  • Mobile first at 360px; the branch supplies no dark mode
  • Wording must avoid library jargon such as 'hold' and 'bib record'
  • Availability data refreshes only every 15 minutes, so states can be stale
  • Must work for members with no email address on file

Deliverables

  • One user flow from search to confirmed reservation
  • Four mobile screens: results, title detail, branch choice, confirmation
  • Empty, unavailable and stale-data states
  • A one-paragraph note on the words you chose instead of 'hold'

Timeline

4 days

Success criteria

  • A first-time member reserves a known title in under 60 seconds
  • Pick-up branch and ready-by date are visible without scrolling on confirmation
  • Nothing in the flow requires understanding the catalogue's structure
IntermediateApp UX / UI2 weeks

Shift swapping for a hospital cleaning team

Verrow Facilities, contracted to two regional hospitals

Problem

Shift swaps happen in a group chat at midnight and are recorded by a supervisor the next morning, which means double-booked shifts and staff turning up for shifts that were given away. Verrow wants swapping inside its own rota app.

Target user

Cleaning staff working rotating nights, many using a shared or older phone, several reading English as a second language.

Scope

Design the request, offer and approval loop for swapping a single shift, including how a supervisor sees and settles conflicts. Payroll, holiday booking and recruitment are out of scope.

Constraints

  • A swap is only valid when both people hold the same certification
  • Supervisor approval is mandatory and can take hours
  • Screens are read in corridors with gloves on: large targets, high contrast
  • Push notifications are unreliable, so no state may depend on one being seen

Deliverables

  • A service flow covering request, accept, approve and decline
  • Six screens at mobile size, including the supervisor's queue
  • A state model for a shift: offered, accepted, pending approval, settled, expired
  • Copy for the four notifications the flow sends

Timeline

2 weeks

Success criteria

  • Every party can tell whose action is awaited at a glance
  • No path allows two people to believe they hold the same shift
  • A swap can be completed without any written message
AdvancedApp UX / UI3 weeks

A control dashboard for a regional energy grid

Torva Grid, an operator balancing wind, solar and battery supply

Problem

Controllers watch six separate tools to decide when to discharge batteries. During a fault, the decision that matters is made under two minutes of pressure with numbers scattered across three monitors, and new controllers take nine months to become confident.

Target user

Trained controllers on twelve-hour shifts at a three-screen desk, plus trainees shadowing them, plus a duty manager checking in from a tablet.

Scope

Design the primary balancing view and the fault-response path within it. Billing, maintenance scheduling and public reporting are out of scope.

Constraints

  • Colour alone can never signal a fault; the room has controllers with colour vision deficiency
  • Data arrives at different rates, from 1 second to 5 minutes, and must be timestamped
  • The layout must fit a 1440px monitor and remain usable on a 1024px tablet
  • No action may be irreversible without a stated confirmation and an undo window

Deliverables

  • An information hierarchy for the balancing view with rationale
  • The desktop view plus the tablet adaptation
  • A fault sequence shown as three timed states
  • An alert taxonomy: severity, sound, colour, shape and wording
  • A short accessibility note covering non-colour signalling

Timeline

3 weeks

Success criteria

  • A controller can name the current risk within five seconds of looking up
  • Every figure on screen shows how old it is
  • A trainee can follow the fault sequence without a colleague narrating it

Turning a brief into a portfolio case study

  1. 01

    Write down your assumptions first

    Before any research, list what you believe about the users and the problem. Dating that list is what later lets you show a change of mind, which is the part reviewers actually read.

  2. 02

    Do small, honest research

    Three or four conversations with people close to the audience beats an invented survey with 200 fictional responses. Say plainly how many people you spoke to.

  3. 03

    Frame the problem in one sentence

    Turn the brief's problem into a sentence you can test a design against. If a screen does not serve that sentence, it does not belong in the case study.

  4. 04

    Explore more than one direction

    Show two or three structurally different attempts, not three colours of the same layout, and say why you dropped the ones you dropped.

  5. 05

    Build the flow, not a poster

    Prioritise the screens that carry the decision, including empty, error and loading states. Those states are what separate a student project from a considered one.

  6. 06

    Test with five people

    A clickable prototype and five short sessions will find most of what is wrong. Record the tasks you set and what changed as a result.

  7. 07

    Close with a reflection

    State what you would do next with more time, and what you would do differently. Note clearly that the brief was a practice assignment.

Questions about UX briefs

What is a UX design brief?
A UX design brief is a short document that sets up a design problem: who the client is, what is going wrong, who the users are, what limits the work, what you are expected to hand over, how long you have and how the result will be judged. It defines the problem rather than describing a solution.
Can I use a fictional UX brief in my portfolio?
Yes, as long as you label it. Say on the case study that the client and brief are fictional practice work. Reviewers care about how you reason, and hiring teams treat undisclosed fake client work as a credibility problem rather than a portfolio gap.
What should a beginner UX project include?
A clear problem statement, the assumptions you started with, a small piece of real research, one complete flow rather than many disconnected screens, the empty and error states, evidence that you tested it with a few people, and a short reflection on what you would change.
How long should a UX portfolio project take?
Long enough to be finished and short enough to end. A first project is reasonable in four to five days, a mid-level one in about two weeks, and a complex system in three. Projects that run for months usually stall before the write-up, which is the part that gets read.

Get a fresh UX brief

Pick a difficulty and get a new fictional client, problem, audience, deliverables and timeline in a few seconds.