User experience (UX) design consulting is advisory work that decides how a person gets something done inside a product or service, then designs the structure, flows and states that let them do it. It starts upstream of the interface, because the questions that determine whether a flow works are business questions wearing design clothing. Vivaldi has worked on brand, product and experience design since 1999, including the user flow redesign behind Rally’s platform relaunch.
- Since 1999brand, product and experience design work
- 200+published case studies
- 68%conversion rate increase for Rally within one month of its relaunch
- 10 officeson four continents
- 5xWirtschaftsWoche Best of Consulting, Strategy and Innovation
Most bad interfaces are not design failures. They are decisions the business never made, arriving at the screen to be resolved by a designer who was not in the room. This page covers where UX work goes wrong, what our UX design consulting involves, how to evaluate any firm doing it, and where a project usually starts.
When companies bring us in
The pattern is consistent. The product works, and users keep describing it as hard.
- A redesign shipped, looked better, and the same complaints came back within two quarters
- Activation or completion rates stalled on one flow that everyone agrees is the important one
- Navigation mirrors the internal org chart, and nobody has the authority to decide which team’s feature comes second
- Support volume concentrates on a single step that exists for a policy reason nobody has revisited
- An AI feature, a personalization layer or a new segment has to be fitted onto a path that was never designed to vary
Why another redesign does not fix it
The reasonable version of the standard argument is that interfaces accumulate cruft, users change, and periodic redesign is maintenance. That is true, and it is not what usually happens.
What usually happens is that a redesign inherits every unresolved argument the organization has had about its product. The result is predictable:
- Navigation that encodes internal structure rather than how users describe the job
- A step that exists only to satisfy a policy nobody has revisited in four years
- Personalization added on top of a fixed path, rather than a path designed to vary
- A design system that is immaculate and a task that still takes eleven clicks
None of that is fixed by better visual craft. It is fixed by settling the decisions before the wireframes, which is slower at the start and considerably cheaper by launch.
The second failure is sequencing. Testing scheduled after the build is not testing, it is a defense of work already paid for. By then, acting on what you learn costs enough that most organizations quietly do not.
What our UX design consulting services include
Four bodies of work, roughly in this order. Which of them you need depends on how settled the underlying product decision already is.
1. Deciding what the interface is for
Before any structure is drawn, the task has to be named precisely.
- Who is doing this, and what state they arrive in: first time, returning, mid-problem, under time pressure
- What counts as done, stated as an outcome the user would recognize rather than a screen they reach
- Which steps exist for the user and which exist for the business, named separately so the trade is deliberate
2. Structuring the flow
Sequence, hierarchy, what is deferred, what is asked for and when. Most of the value in a UX engagement is created here, and it is the part most often skipped in favor of screens.
- Information architecture and navigation built from how users describe the job, not from the product taxonomy
- Flows for the two or three highest-value tasks, designed to completion including the unhappy paths
- Explicit decisions about what to remove, the hardest deliverable to get signed off and usually the most valuable
3. Designing for people and for software
An interface increasingly has to be legible to software as well as to people. Assistive technology has always required this, and automated agents acting on a user’s behalf now require it too.
- Accessibility treated as a structural requirement, not a remediation pass at the end
- Clear semantics, labeling and state, so a screen reader or an agent can interpret what a sighted human infers from layout
- Content and interface copy designed with the flow, because interfaces are mostly language
4. Testing before it is expensive to change
- Prototypes at the lowest fidelity that can answer the question honestly, tested with real users while changing course is still cheap
- A measure agreed in advance, whether that is task completion, time to first value or conversion on the flow, so the argument after launch is about evidence
This page previously published a point of view on synthetic customer experiences, and our position on simulated users has not changed: they can narrow an option set cheaply, which is a real use, and they are weakest exactly where the stakes are highest. They lower the cost of being wrong early. They do not remove the need to be right later.
- DefineWhat the task actually isWho it is for, what counts as done, and which steps exist only for the businessUsually assumed
- StructureThe shape of the flowSequence, hierarchy, what is deferred, and what gets removedUsually thin
- TestWhether it holds with real usersLow-fidelity prototypes, run while changing your mind is still cheapUsually too late
- BuildScreens, components and shipWhere the budget, the craft and the attention already areUsually strong
Our work here is usually commissioned by a Chief Product Officer, a Head of Digital or a founder past first product-market fit. A defined flow typically runs six to twelve weeks, longer where the underlying product decisions are still open.
Not sure which of these four you need first?
Start with a working session. We will take one flow, establish what it is supposed to accomplish, and name the decisions inside it that nobody has made yet.
How a UX design engagement works
- Frame. Interviews with the people who own the product and the people who support it, a review of the analytics and support data, and a walk through the current flow as a user meets it. The output is a short list of tasks worth designing for, and the decisions blocking them.
- Structure. Information architecture, navigation and the flows for the highest-value tasks, including errors and edge states. The output is a decision about what is asked, in what order, and what is removed.
- Test. Prototypes at the fidelity the open question requires, put in front of real users, with the measure of success agreed before the session rather than after it.
- Hand over. Specifications a build team can work from without further interpretation, the patterns that govern the next screen nobody has drawn yet, and a named owner for the exceptions that will follow.
Where this has been done
Rally, a bus-sharing platform, had built a business on crowd power: anyone could create, sign up for or share a trip, and enough interest in one direction put a bus on the road. The commercial problem was that the platform’s own flow made those rider actions harder than they needed to be.
Vivaldi audited Rally’s existing internal structure to find where the friction sat, studied how other crowd-powered companies inside and outside transportation engaged their audiences, and redesigned the core user flow. Through workshops and interviews we developed an algorithm and service design for connecting riders traveling on the same trip, then produced a comprehensive sitemap with wireframes for the app, desktop and tablet experiences, plus the design and messaging direction and core page designs that carried the brand through them.
Rally relaunched with a 68% conversion rate increase within one month. Read the full story: Adding fuel to a platform business with crowd power.
Why companies choose Vivaldi for UX design work
We are a strategy firm that designs, which cuts both ways and is worth being clear about. If you need a large interface production capability, there are better options than us. If the hard part is that nobody has decided what the product is supposed to do for whom, that is the work we do first and the reason the design holds afterwards.
Three practical consequences. The product decision and the interface decision get made in the same room, so a flow is not redesigned in comments two weeks after sign-off. The structure of the task is treated as the deliverable, not as a stage on the way to screens. And we will tell you when the answer is to remove a feature rather than to redesign it, which is the recommendation an agency paid by the screen has no incentive to make.
How to evaluate a UX design consultant
Including us. These are the questions worth putting to every firm on your shortlist.
- Ask what they will refuse to design. A firm that will draw whatever is on the backlog is selling capacity. The useful answer names a step they would push back on, and why.
- Ask when they test. If user testing appears after the build in their plan, the plan is a validation exercise. Testing has to sit where changing your mind is still cheap.
- Ask who writes the words. Interfaces are mostly language. If copy arrives from a separate team after the screens are approved, the flow gets redesigned in comments.
- Ask how they handle accessibility. If the answer is an audit near the end, expect rework. If the answer sits in the component and semantic decisions, they have done this before.
- Ask what happens after handoff. Who owns the pattern library, who arbitrates the next exception, and whether the honest answer is you or them.
UX design consultancy vs digital design agency vs in-house product team?
| Criterion | UX design consultancy | Digital design agency | In-house product team |
|---|---|---|---|
| What it decides | Which tasks the product is for, the structure of the flow, and what gets removed | How the flow is expressed in screens, components and visual language | What ships next quarter, within the structure the product already has |
| Typical deliverables | Task definition, information architecture, flows including edge states, tested prototypes, build-ready specifications | Screen designs, design system, production assets | Backlog, releases, incremental optimization |
| Who signs it off | Product and commercial leadership together, because the trade-offs are commercial | Design or marketing leadership | Product leadership |
| Blind spot | Limited production throughput once the structure is settled | Will render a flow with the wrong shape very beautifully | Cannot easily question a structure it is measured on shipping within |
| Best when | A redesign has already failed, or the product decision underneath the interface is still open | The flows are settled and the expression is tired | The structure is sound and the work is continuous improvement |
Related work
UX design rarely sits on its own. Four connected pieces, depending on where your question actually is.
- Customer ecosystem strategy and interaction field design if the question is the whole system of participants around the brand rather than one product’s interface.
- User interface (UI) design if the flows are settled and the work is the visual and component layer.
- Service design if the experience runs across human and digital touchpoints rather than inside a screen.
- Prototyping and testing if what you need first is evidence about a concept, before anyone builds it.
This work sits inside our Experience Strategy practice, part of the argument our Brand, Innovation and Experience pillar makes: an experience is where brand meaning either survives contact with reality or does not. It is delivered at Vivaldi, the strategy firm behind the Demand Landscape and the Interaction Field.
UX design consulting in the US and worldwide
Our experience design work runs out of ten offices on four continents, with the US team in New York. Interfaces are where multi-market strategy gets tested most bluntly: the same flow meets different payment habits, different regulatory steps and different assumptions about what a form is allowed to ask. We design for the markets you actually operate in rather than localizing a US flow afterwards, because the step that has to change is usually structural rather than linguistic.
Start with the one task that matters most
A working session on a single flow: what it is supposed to accomplish, how users describe it, and the decisions inside it that nobody has made yet. You leave with the shape of the problem and where to start.
