A dedicated designer embedded in your team on a monthly subscription, two-month minimum, with a trial first.
Top 12 SaaS product
design agencies for 2026
Studios that design the inside of SaaS products, from dashboards to
permissions, with engagement models, pricing and honest trade-offs.
This page covers in-app design, not marketing sites. The distinction matters because the work is different in kind. Nobody arrives at your product to be persuaded. They arrive to do a job, often for the fourth time today, and the design question is how fast they can do it and how confident they feel doing it.
That shifts the emphasis onto the screens that never make a portfolio: permissions and roles, settings, billing, bulk actions, empty states, the admin panel your support team lives in, and the dashboard everyone asked for and nobody can read. The studios below have shipped that work. Website specialists and brand studios are covered elsewhere on this site.
Best for
Start from the bottleneck
Six situations cover most in-app design engagements. Pick the one that sounds like your product and the shortlist narrows to two or three names.
Product design is rarely finite. When the roadmap has no end date, buying a series of projects means re-briefing a studio every quarter and paying for ramp-up each time.
A hard cap on concurrent engagements, so the principals who scope the work stay on it throughout.
The hardest structural problem in B2B software, and the one most often deferred: an admin, a manager and an end user all needing different versions of the same screen.
Enterprise interface architecture for functionally heavy platforms, with design systems as a core service rather than an upsell.
Dense internal interfaces across workflow, healthcare, logistics and HR tech, with components delivered in code.
You are adding an assistant or model-driven features and nobody has settled the interaction pattern yet, or the data itself is the interface problem.
A long focus on AI experiences rather than a recent pivot to them, for AI, data and B2B platforms.
Strategy, research and design systems for B2B SaaS, AI and digital health products.
Nobody can agree what the product should be, so the problem has to be defined before the interface is.
Every design proposal keeps running into what the backend will and will not allow, so the decision has to be made with the engineers in the room.
In-house development, so components arrive in code rather than specified for someone else to interpret.
Design and development running together rather than sequentially, for workflow-heavy B2B products.
Design decisions made alongside the engineers who own and maintain the platform.
The interface itself is the competitive argument, or you have a behavioural question that guesswork has failed to answer.
The agencies
Twelve worth knowing
All twelve have shipped in-app work on real products. Each entry states how the studio works and what would make it the wrong choice.
01
Eleken
Best for: products that grew features faster than logic
SaaS product design as the entire business, sold by the month.
What they do: In-app UI/UX, product redesign, UX audits, design systems, and prototyping for SaaS. No development, no marketing sites, no brand work.
Proof: More than 200 SaaS products, including Gamaya in precision agriculture, TenderX in freight tendering, and TextMagic.
How they work: A dedicated designer embedded in your team on a monthly subscription, two-month minimum, with a trial first. Its published thinking on untangling products that outgrew their structure is the clearest signal of what it actually does.
Watch for: You supply the engineers. And a rolling model is poor value if you have one finite, well-specified redesign.
eleken.co- Base
- Ukraine and remote, mostly GMT+2
- Founded
- 2015
- Scale
- 100-plus people
- Not offered
- Development, marketing sites, brand
02
Clay Global
Best for: multi-role products where the question is who sees what
Dense multi-role software, where the design question is who sees what.
What they do: Enterprise and SaaS product UX and UI, design systems, UX research, and interface architecture for functionally heavy platforms.
Proof: Product and interface work for Slack, Stripe, Coinbase, Google, Amazon, and Sony.
How they work: Research-led project engagements, with design systems as a core service rather than an upsell, which matters when your team ships features for years afterwards.
Watch for: Premium positioning, and the studio's breadth across brand and web means you are paying for capability you may not need on a purely in-app engagement.
clay.global- Base
- Remote-first, San Francisco HQ
- Office
- New York, 148 Lafayette St
- Model
- Research-led project engagements
- Ownership
- Independent
03
MetaLab
Best for: a significant product bet where the interface is the argument
Subtraction as a discipline, on products people use daily.
What they do: Product design, UI/UX, prototyping, design systems, and engineering.
Proof: Early interface work for Slack, Coinbase, and Uber Eats, plus product work for Apple, Disney, and Google.
How they work: Senior-led project engagements running from strategy through build, with a house approach built on removing elements rather than accommodating every stakeholder request.
Watch for: Price and availability both reflect the reputation, and it is oversized for incremental feature design.
metalab.com- Base
- Victoria BC, remote-first across roughly 19 countries
- Founded
- 2006
- Model
- Senior-led, strategy through build
- Ownership
- Independent
04
Lazarev.agency
Best for: assistants and model-driven features with no settled pattern
AI features and dense data, designed as interaction problems.
What they do: Research, product design, UI, and design systems for AI, data, and B2B platforms.
Proof: Boeing and Accern, plus a UX redesign for Peel that preceded its acquisition by Shopify.
How they work: Project engagements covering research through system delivery, with a long focus on AI experiences rather than a recent pivot to them.
Watch for: The portfolio spans marketing sites and brand work too, so it is broader than an in-app specialist, and US rates sit above the European studios here.
lazarev.agency- Base
- San Francisco
- Scale
- Roughly 50 people
- Working since
- 2017
- Ownership
- Independent
05
Cieden
Best for: operational complexity that has to survive the data model
Dense internal interfaces, with engineers in the same team.
What they do: Product design, UX audits, and design systems across enterprise workflow, healthcare, logistics, and HR tech, with in-house development.
Proof: MedEntry, Sidekick, Hygrid Health, and Lykon, with 45-plus healthcare projects reported.
How they work: Project engagements, frequently audit-first, with components delivered in code rather than specified for someone else to interpret. It publishes an enterprise app design handbook covering its approach to heavy interfaces.
Watch for: Seven countries means US clients should plan overlap deliberately, and client feedback notes that mid-project scope additions raise costs.
cieden.com- Base
- USA, Canada, Ukraine, Czechia, Norway, Spain, Romania
- Scale
- Roughly 47 people
- Published rate
- $50 to $99 per hour, projects from $10,000
- Development
- In-house
06
Mission Control
Best for: difficulty that sits in the logic rather than the surface
Two senior designers on the hard part, and nothing else.
What they do: Product and UX design, UI, and design strategy for complex software at startups and growth-stage companies.
Proof: No public named-client list, so judge this one from the portfolio and reference calls directly.
How they work: Fully remote and async-first, with a hard cap on concurrent engagements so the principals who scope the work stay on it throughout.
Watch for: Limited public proof and small capacity. It cannot run alongside a large parallel programme.
missioncontrol.co- Base
- Fully remote, async-first
- Model
- Capped concurrent engagements
- Public proof
- None published
- Ownership
- Independent
07
Qubstudio
Best for: teams that arrive without a settled brief
Nineteen years of B2B product work, with a documented process.
What they do: Research, UX/UI, product management, strategy, and design systems across fintech, SaaS, healthcare, and logistics.
Proof: More than 600 projects, including Gulf Bank, ila Bank, the BluePallet industrial commerce platform, Loan Mantra, and Candesic.
How they work: Workshop-led discovery followed by staged delivery, with a published design system method running from audit through adoption planning.
Watch for: No in-house development, so a build partner is a separate arrangement, and reference work sits mostly in Europe and MENA.
qubstudio.com- Base
- UK, Ukraine and Spain
- Scale
- Roughly 54 to 65 people
- Published rate
- $50 to $99 per hour
- Not offered
- In-house development
08
Merge
Best for: workflow-heavy B2B products without strong front-end capability
Workflow-driven B2B products, with build attached.
What they do: Product and UX design, UI, and product development for B2B, SaaS, and fintech.
Proof: SaaS and fintech design and development spanning early MVPs through established platforms, without a prominently published named-client list.
How they work: Project engagements with design and development running together rather than sequentially.
Watch for: Thin public proof relative to others here, so more assessment falls on your own reference calls.
merge.rocks- Base
- Distributed
- Focus
- B2B, SaaS and fintech
- Development
- In-house, alongside design
- Ownership
- Independent
09
925Studios
Best for: regulated categories where compliance shapes the interface
Craft plus compliance awareness, for funded product teams.
What they do: Strategy, research, product design, and design systems for B2B SaaS, AI, and digital health products.
Proof: Design systems, apps, and platforms for Series A to C healthtech companies across digital therapeutics, patient engagement, and provider tools.
How they work: Project engagements with founders and product teams, building accessibility and regulatory expectations into the design rather than retrofitting them.
Watch for: Founded 2022, so the track record is shorter, and the weighting toward healthtech means fewer large multi-product engagements.
925studios.co- Base
- Distributed
- Founded
- 2022
- Focus
- B2B SaaS, AI and digital health
- Ownership
- Independent
10
DOOR3
Best for: software where customers complain and nobody can name the cause
Products with a usability problem nobody can name.
What they do: UX audits, research, product design, design systems, and data visualization for complex applications.
Proof: Cadwalader's Connect platform, DAA Enterprises pharmacy management software, Surex Direct, Aptivio, and a digital overhaul for NJIT.
How they work: Consultancy engagements that frequently open with a UX audit before any redesign gets scoped, which suits products where the complaint is vague but persistent.
Watch for: A technology consultancy rather than a design studio, so visual craft is not what you are buying.
door3.com- Base
- New York City HQ, global team
- Founded
- 2002
- Model
- Consultancy, audit-first
- Ownership
- Independent
11
ustwo
Best for: a behavioural question guesswork has failed to answer
Research that answers behavioural questions properly.
What they do: UX research, interaction design, service design, UI, and product strategy across enterprise, financial, and medical products.
Proof: DeepMind Health's Streams app for clinicians, plus product work for Google, HSBC, Peloton, and Samsung.
How they work: Project engagements with a genuinely staffed research function rather than a discovery phase bolted onto design.
Watch for: SaaS is one segment among several rather than a dedicated practice, and it is priced as a premium studio.
ustwo.com- Base
- London HQ, Brooklyn studio
- Also in
- Malmo, Lisbon, Tokyo
- Founded
- 2004
- Status
- Independent, B Corp
12
Bitovi
Best for: design proposals that keep hitting backend limits
For when the honest constraint is architectural.
What they do: Systems engineering, frontend, AI, design, and program management for large software organisations.
Proof: Enterprise software work for clients including Yum! and Coinbase.
How they work: Consultancy engagements where design decisions get made alongside the engineers who own and maintain the platform.
Watch for: Engineering-led, so research depth is not the strength and design arrives inside a broader technical engagement.
bitovi.com- Base
- Distributed, US-based
- Model
- Consultancy, engineering-led
- Also covers
- Systems engineering, frontend, AI
- Ownership
- Independent
Compare
The same facts, side by side
Stated facts only: where each studio is based, how it sells the work, whether development sits in the same company, and what it publishes about rates.
| Studio | Base | How they sell the work | Development | Published rate |
|---|---|---|---|---|
| Eleken |
Ukraine, remote | Monthly subscription, embedded | None | Not published |
| Clay Global |
San Francisco, New York | Project, research-led | Not stated | Not published |
| MetaLab |
Victoria BC, remote-first | Project, senior-led | In-house | Not published |
| Lazarev.agency |
San Francisco | Project | Not stated | Not published |
| Cieden |
Seven countries | Project, often audit-first | In-house | $50 to $99 per hour, from $10,000 |
| Mission Control |
Fully remote, async-first | Embedded, capped engagements | Not stated | Not published |
| Qubstudio |
UK, Ukraine, Spain | Workshop-led, staged delivery | None | $50 to $99 per hour |
| Merge |
Distributed | Project, design plus build | In-house | Not published |
| 925Studios |
Distributed | Project | Not stated | Not published |
| DOOR3 |
New York City | Consultancy, audit-first | Not stated | Not published |
| ustwo |
London, Brooklyn, plus three | Project, research-led | Not stated | Not published |
| Bitovi |
Distributed, US | Consultancy, engineering-led | In-house | Not published |
Bought as capacity
People inside your team for as long as the roadmap runs, priced by the month rather than the project.
Bought as diagnosis
The engagement opens with an audit or a workshop, because the problem is not agreed yet.
Bought as craft
Interface quality is the deliverable, and the studio is priced accordingly.
Ask to see these
The screens that decide whether a redesign worked
Portfolios show dashboards and onboarding. Renewals turn on other things entirely, so ask to see these before you sign.
Permissions and roles
The hardest structural problem in B2B software, and the one most often deferred. Ask a candidate studio how they handled a product where an admin, a manager, and an end user needed different versions of the same screen.
Settings
Usually a decade of accumulated toggles with no organising logic. Nobody wants to design it, everybody has to use it, and a good studio will have opinions about how it should be structured.
Empty states
What your product looks like on day one, before any data exists, which is exactly when a new customer decides whether it was worth buying. Frequently designed last, if at all.
Bulk actions and data tables
Where real users spend their time in operational products. Sorting, filtering, selecting fifty rows and doing something to them. Rarely in a case study, always in daily use.
The admin panel
Your support and customer success teams use it constantly, and its quality directly determines your cost to serve. It gets designed by engineers by default.
Redesigns
Redesigning a product people already use
Greenfield SaaS design is rare. Most engagements are redesigns, and they fail differently.
Existing users have built muscle memory on your current interface, including its flaws. A change that tests beautifully with new users can still infuriate the customers paying your bills, and the churn shows up a quarter later when nobody connects it to the redesign.
What works is transitional. Stage the rollout rather than switching everyone at once. Give power users an opt-in beta and take their complaints seriously, because they are your renewal base. Keep changed elements in recognisable positions even when the behaviour improves. Run both interfaces in parallel for a period if the architecture allows it.
Ask candidate studios directly how they handled migration on a previous redesign. Anyone describing launch day as a single switch has not done this on a product with real customers.
FAQ
Questions buyers actually ask
Project, retainer, or subscription for in-app work?
Product design is rarely finite, which is why subscription and retainer models have spread through this category faster than any other. A redesign or a new module is project-shaped. Continuous feature design is not, and buying it as a series of projects means re-briefing a studio every quarter and paying for ramp-up each time. Match the model to whether your roadmap has an end date.
Do we need a studio that also builds?
It depends on your front-end capacity, not on principle. If you have engineers, a design-only partner like Eleken avoids paying for a build team you do not need. If your interface depends on technical decisions nobody has made, or your engineers are backend-weighted, a combined team resolves that in place instead of across a specification. The common failure is hiring full-service when you needed two designers.
How do we brief a studio on a product this complex?
Give them access rather than documentation. A domain walkthrough with someone who actually uses the product, recordings of support calls, and your analytics will do more in a week than a specification does in a month. Studios experienced in B2B will ask for exactly that. Ones that accept a written brief and disappear are about to design from your assumptions rather than your customers' behaviour.
What should this cost?
European studios here run $50 to $99 per hour with project minimums from around $10,000. US studios sit materially higher, and the premium names higher again. Subscription models are priced per designer per month and usually work out cheaper past six months. Price the same scope in both structures before deciding, since the packaging changes the total more than the rate does.
How do we measure whether it worked?
Agree the measures before the work starts and take a baseline. For in-app design the useful ones are task completion time, error and rework rates, support ticket volume on the redesigned area, and feature adoption. Satisfaction scores move for reasons unrelated to design. If a studio cannot propose a measurement plan, it has not been held to one before.
How long does it take?
Eight to sixteen weeks for research and design on a defined product area or a major redesign. Twelve to twenty weeks if a design system is part of the scope. Build discovery in rather than skipping to design, because in B2B the domain complexity is almost always what catches teams out, and time saved in week one gets spent twice in week six.