The shape of the week
A typical UX designer's week splits across three broad categories. Synthesis — turning research, analytics and stakeholder input into structured problem framings. Production — making flows, wireframes, prototypes, microcopy and the artefacts engineers build from. Decision work — sitting in meetings, defending recommendations, communicating risk, revising as needed.
For a mid-level designer, the split is roughly even — about a third in each. For a senior designer, synthesis and decision work grow to dominate; production shrinks. For a junior designer, production dominates; synthesis is supported and decision work is mostly observed. This is the underlying reality the rest of this article is unpacking.
This is the operational answer to "what does a UX designer do". It is also the answer to why senior UX work pays more — the work shifts from making screens to making decisions, and the decisions compound. The broader Fundamentals frame: what is UX design and the UX design process.
Research work
Most UX designers do some research, but the depth varies sharply with team size. In a small product team — under ten people, one or two designers — the UX designer runs interviews, plans usability tests, recruits participants and synthesises findings themselves. The cycle is short, the methods are lightweight, and the same person carries the work from question to recommendation.
In a larger organisation with a dedicated research function, the UX designer is more often a research consumer than a researcher. They shape questions in collaboration with researchers, attend selected sessions, interpret findings, and translate them into design decisions. The hands-on execution sits with the research team; the design implications sit with the designer.
Both shapes require research literacy — knowing what each method costs and returns, when to reach for each, what evidence is needed to inform which decision. The depth of execution skill varies. Designers in small teams get good at execution fast; designers in larger orgs get good at interpretation. The detailed reference: UX research methods and the specific operational guides for user interviews and usability testing.
Flows and design production
The visible part of the work. Flows — what happens when a user clicks here, what state the system enters, what the screen shows next. Wireframes — low-fidelity structural layouts that explore the shape of a solution. Prototypes — interactive Figma files that test the experience before code is written. Production design — high-fidelity, component-accurate, ready-to-build screens.
For most working UX designers in 2026, Figma is the default tool for all of this. Strong fluency in components, auto-layout, variables and design tokens is assumed at mid-level. Production design takes longer than it looks because the work is in the detail — edge cases, empty states, error states, loading patterns, accessibility considerations, content accuracy. A senior designer's spec is usually 30–40 per cent edge-case coverage; a junior designer's spec is mostly the happy path.
AI has compressed the first-pass production work meaningfully. A flow that took two days to wire up in 2022 can be drafted in an hour in 2026; the designer's time is now spent reviewing, correcting and polishing rather than drawing from scratch. The judgement layer — what should the flow do, why, for whom — hasn't compressed.
Workshops and stakeholder alignment
Workshops are not a separate UX activity; they're a tool used in service of the synthesis and decision work. A working UX designer runs workshops in three common situations. Discovery, where the team needs to align on what the problem is. Prioritisation, where the team needs to decide what to build. Review, where the team needs to give and receive design critique.
The structure of a good workshop is much less important than the framing of the question it answers. A bad workshop produces a wall of sticky notes that nobody references again; a good workshop produces a one-paragraph decision the team can act on. The difference is in the preparation, not the facilitation. Senior UX designers spend more time on workshop preparation than on running the session itself.
Outside of workshops, stakeholder alignment happens in smaller conversations — one-to-ones with the PM, syncs with engineering leads, status updates to executives. This work is largely invisible from the outside but absorbs a meaningful share of the senior designer's time. It is the work that gets a design through review without losing the thing that made it good.
The hours spent in stakeholder conversations don't appear in the portfolio. The decisions they produce are the difference between a senior practitioner and a mid one. Designers who treat stakeholder communication as a tax on the "real" design work plateau in their careers; designers who treat it as the high-leverage half of the work get promoted.
Testing and validation
Validation is part of the role, but in three quite different forms.
Usability testing — sessions with real users on a prototype or live product, observing where they struggle. Most working UX designers run two to six sessions a month, depending on the project cycle. The output is qualitative findings translated into design changes. The detailed operational view: usability testing guide.
Internal review — design critiques with colleagues, presentations to stakeholders, walkthroughs with engineering. Less rigorous than user testing but quicker, and useful for catching obvious problems before they go to users.
Post-launch validation — measuring whether the shipped design actually improved the metric it was supposed to. Often skipped by junior designers; central to senior designers' credibility. A designer who can say "we shipped this, here's what happened, here's what we'd do next" has evidence of commercial impact; a designer who can only point to artefacts does not.
Prioritisation and trade-offs
UX designers don't usually own the product roadmap. Product owns it. But UX designers participate in prioritisation through evidence, advocacy and the quality of their arguments. The work is: bringing user evidence into the prioritisation conversation, naming the UX risk of de-prioritising something, and recommending what to cut when scope has to shrink.
Strong UX practitioners learn to argue for cutting features. The junior instinct is to fight for more scope — more research, more design, more polish. The senior instinct is to argue for less of what doesn't matter so there's room for more of what does. The discipline isn't to win every argument; it's to be the person in the room making the best version of the case, every time.
Documentation and handoff
Documentation absorbs more time than juniors realise. Annotating designs for engineering. Writing decision memos. Producing handoff specs. Updating tickets in the project management tool. Keeping the design system in step with what's being shipped. None of this is glamorous; all of it determines whether the design actually ships in the form it was designed in.
Mature teams have lightweight documentation practices — short decision memos rather than long specs, in-Figma annotations rather than separate PDFs, conversations in the project management tool rather than long email chains. The discipline is to write enough that an engineer can build from it without you in the room, but not so much that nobody reads it.
Product collaboration
UX designers spend more time with their PM than with any other colleague. The relationship is the single biggest predictor of whether the UX work gets shipped well. A strong UX-PM relationship looks like: shared problem framing before any design happens, weekly syncs on what's coming, the designer being involved in roadmap conversations before the brief lands.
The other consistent collaboration is with engineering. A senior UX designer treats engineering as a partner in problem-solving — engineers know things about feasibility, performance and edge cases that designers should learn from. Designers who design in isolation and toss specs over the wall produce work that gets reshaped in the build; designers who pair early produce work that ships closer to intent.
Other collaborators vary by org. Research, content design, accessibility specialists, design system maintainers, brand and marketing in some companies. The pattern: senior UX designers have a wider network of internal collaborators than juniors do, and that network is part of why they ship better.
Junior vs senior responsibilities
The shift from junior to senior UX is not "more of the same work, better executed". It's a different category of work.
What changes from junior to senior
Junior (0–3 years). Mostly production. Wireframes, flows, prototypes, content from briefs that someone else framed. Light synthesis. Decision work observed rather than driven. Internal review is the main feedback loop.
Mid-level (3–5 years). Production still significant but synthesis and decision work growing. Designers at this level run their own research-to-design cycles, present to stakeholders, defend recommendations. The transition years; some designers plateau here.
Senior (5+ years). Mostly framing and decision. Production reduced. The work is choosing what to build, defending the choice, communicating risk, mentoring more junior designers. Senior designers often have a specialism — research, IA, design systems, interaction — that anchors their craft.
Lead and above. Less hands-on work. People management, design strategy, cross-team coordination, hiring. The work shifts from doing design to enabling others to design well. Some senior designers don't want this transition; the IC path (principal, staff) is increasingly viable.
The transition usually happens between years three and five, but it is driven by behaviour rather than tenure. Designers who learn to frame problems, defend recommendations and communicate to stakeholders move up. Designers who only get faster at production stay at mid-level. The career-path detail: senior vs junior UX designer.
Startup vs agency vs enterprise
The same role looks materially different depending on where it's held.
Startup (under 50 people). Wide scope, short cycles, fast decisions. One UX designer often holds research, UI, content and front-end engineering pairing. Lower formal process, higher individual visibility, more risk and reward. Strong fit for designers comfortable with ambiguity. Best for early-career designers who want to learn fast; harder for designers who want defined remits.
Scale-up (50–500 people). The most common shape in the UK in 2026. Teams of two to fifteen designers. UX work is real and respected; processes are emerging but lightweight. Most career growth happens here. Pay is mid-range; equity often part of the package.
Enterprise (500+ people). Specialised roles. Slower cycles, more political work, longer projects. UX research, content design, accessibility, design systems often separate functions. Best for designers who want depth in a specialism; harder for designers who want broad exposure.
Agency or consultancy. Multiple client projects in parallel. Cycles are bursty — intense delivery periods, calmer prep weeks. Exposure to many domains and team shapes; depth in any one product is shallower. Strong route for designers who want to become commercially literate fast; harder for designers who want to see long-term outcomes of their work.
The honest career advice: most designers do well to work in two different shapes during their first ten years. An agency stint plus an in-house stint produces a stronger practitioner than a decade of either alone.
What UX designers don't do
Three areas commonly misattributed to UX.
Marketing copy. UX designers write microcopy — error messages, empty states, button labels, the words that live inside the product. They do not typically write marketing campaigns, landing-page hero copy or social-media content. That's marketing or brand work. The boundary moves in small teams; in mature orgs it's clearer.
Brand design. Logo, identity, brand-level visual language. Sits with brand or marketing design, not UX. UX designers work within the brand system; they don't usually create it.
Full front-end engineering implementation. UX designers specify the design, partner with engineering on implementation details, sometimes pair on QA. They don't typically write production code. The line moves in small teams (where one person sometimes designs and builds) and at the senior end (where some specialists work between design and engineering as design engineers).
The honest summary: UX designers do less than the most expansive job descriptions suggest. The role is broad but not infinite. Roles that try to combine UX, brand, marketing, content and full-stack engineering are wishlists from teams that haven't decided what they need; candidates who recognise the pattern interview with that awareness.
How AI has changed the work
By 2026, AI has reshaped what a UX designer's week looks like. The pattern is consistent across teams: AI accelerates the mechanical layer of the work; the judgement layer remains human.
What AI now does fast in 2026 that it didn't in 2022: drafting first-pass copy, summarising research transcripts, clustering interview themes, generating flow scaffolds from prompts, producing asset variants, writing decision memo drafts, code-friendly handoff annotations. Tasks that absorbed days now absorb hours. The implication: designers can run more research cycles in the same calendar time, ship more design work, and operate at a level of leverage that wasn't available three years ago.
What AI does not do: recognise what an interviewee isn't saying. Decide which user matters most when their needs conflict. Defend a design choice to a sceptical executive. Frame a problem so the team can actually solve it. Know when the stakeholders are asking the wrong question. The framework anchor: what AI should not replace in UX.
The career-side implication: junior UX hiring has softened because production is the part AI does cheapest. Senior UX hiring has held or grown because direction and judgement are the parts AI hasn't compressed. Designers in 2026 are doing less production and more direction; the role has gained leverage rather than been replaced. The broader operational view: AI for UX designers.
Frequently asked questions
What does a UX designer do day to day?
The week splits across synthesis (turning research and stakeholder input into framings), production (flows, wireframes, prototypes, content), and decision work (defending recommendations, communicating risk, revising). Mid-level designers spend roughly a third of their time in each; senior designers spend more on synthesis and decision, less on production.
Do UX designers do research?
Yes, with depth depending on team size. In small teams, the UX designer runs interviews, plans usability tests and synthesises themselves. In larger orgs with dedicated research functions, the UX designer is a research consumer — shaping questions, interpreting findings, translating to design. Both versions require research literacy; only the first requires execution skill.
How much of UX design is talking to users?
Less than people expect. A working UX designer runs two to six user-facing sessions a month — four to ten hours of contact time and twenty to forty hours of synthesis and design around it. Proportion is higher in research-specialist roles and lower in design-system or interaction-design specialist roles.
What does a UX designer not do?
Marketing copy (that's marketing). Brand design (that's brand). Full front-end engineering implementation (that's engineering). UX writes microcopy, works within brand systems, and partners with engineering on implementation. The line moves in small teams; in mature orgs the boundaries are clearer.
What's the difference between junior and senior UX work?
Junior work is mostly production with light synthesis. Senior work is mostly framing and decision. The transition happens between years three and five and is driven by behaviour (learning to frame, defend, communicate) more than tenure. Designers who only get faster at production stay at mid-level.
What does a UX designer do in a startup vs a big company?
Startup: wide scope, short cycles, multiple hats. Enterprise: specialist role, slower cycles, more political. Agency: multiple clients in parallel, bursty cycles, broad exposure. Each shape has different commercial and learning trade-offs; most strong practitioners have worked in at least two shapes during their first decade.
How has AI changed what a UX designer does?
AI has compressed the mechanical layer — first-draft copy, transcript synthesis, asset generation, flow scaffolds. Senior judgement (framing, decision, defence, recognising what's missing) is unchanged. The role in 2026 is less production and more direction. Junior hiring has softened; senior hiring has held or grown.