Most best agency lists are sorted by who paid to appear on them. This one is not, so let me be direct about my position: I do product design, and I am on this list. Read the rest with that in mind, and judge each entry on whether the description is fair rather than on whether the author is neutral. That is a more useful test anyway. Product design is the broad lens here. And if you haven't settled whether hiring is the right move yet, the six signals that tell you it's time to hire a product designer is worth reading first. If you care specifically about research and testing rigor, the broader UX design agency roundup ranks shops on exactly that.
A note on vocabulary before the list. Agency, studio, firm, company, consultancy. These words mean nothing. They are marketing choices, not categories, and a two person operation calling itself a firm is doing exactly what a forty person operation calling itself a studio is doing. Ignore the noun and look at the structure underneath it.
What actually separates them
Three things, and none of them appear on a website.
Who touches the file. At most agencies above a certain size, the senior person who sells the work is not the person who does it. That is not dishonest, it is how the model works. But it does mean the portfolio you fell in love with may have been made by someone who will not be on your project. Ask directly, and ask to talk to them.
Whether they can ship code. Design that stops at a Figma file gets reinterpreted on the way to production, and the parts that get quietly simplified are always the hard ones you were paying for. Some shops close that gap. Most do not.
Whether they can say no. This is the underrated one. Good product design is mostly deciding what not to build, and a shop that has never talked a client out of a feature is a shop that draws whatever it is handed.
Pavle Lucic
Best for: teams that want the senior designer to also be the person writing the front end.
I design product interfaces and then build them in React and TypeScript, in your codebase. No handoff, no translation layer, no version of the design that got simplified because it was inconvenient to implement.
The work skews technical and dense: dashboards that carry real data volume, design systems where tokens flow from Figma into production without drifting, and AI products where the design problem is trust rather than layout.
The honest limit: I am one person. If you need five people working in parallel, I will tell you so on the first call rather than take the work. Most engagements start with a fixed price UX audit, which is a cheap way to find out whether the fit is real before either of us commits.
IDEO
Best for: organizations that need a strategy answer, not a screen.
The name everyone knows, and the reason design thinking is a phrase your CEO has heard. Expensive, research heavy, and genuinely good at reframing a problem when the client has been asking the wrong question for two years. Overkill if you know what you need built.
Clay
Best for: funded companies who want a recognized partner.
San Francisco, strong portfolio across enterprise and consumer, and increasingly known for AI product work. Polished and priced accordingly. A good fit when the brand of your design partner matters to your board.
Ramotion
Best for: brand and product in one engagement.
Unusually strong at the intersection of visual identity and product interface. If your product needs both a brand and a design system, and you would rather not manage two vendors who blame each other, this shape solves a real problem.
Eleken
Best for: SaaS teams with a steady flow of design work.
Focused on SaaS specifically, with subscription pricing rather than project scoping. Good when you have continuous design need and do not want to renegotiate scope every six weeks. Less good when you need one deep, weird problem solved brilliantly. If SaaS is the whole of your problem rather than one line on it, the best SaaS design agencies is a narrower, deeper list worth checking too.
Fuselab Creative
Best for: data heavy products.
Real data visualization capability, which is rarer than it sounds. Most design shops treat charts as decoration. If your product's job is to make a number mean something, that difference is the whole engagement.
Lazarev
Best for: complex domains nobody on your team can explain simply.
Research led, patient in discovery. If your product lives somewhere genuinely complicated, and the last designer you hired drew beautiful screens that missed the point, this is the failure mode they are built to prevent.
Pencil and Paper
Best for: teams who want to learn the reasoning, not just receive the file.
Focused on complex applications, and unusually generous with public writing about their patterns. That public track record is itself a form of evidence, and it is worth something when your internal team has to maintain the work afterwards.
Netguru
Best for: design and engineering staffed together, at scale.
A large European firm that pairs design with real development capacity. When the problem is that you need both, and you need them coordinated, and your budget supports a full team, this shape is the answer.
The independents worth knowing
Plenty of the best product designers do not work at agencies, and for a lot of teams that is the better answer.
Brad Frost is the reference point for design systems. He named atomic design, and the vocabulary he introduced is now what most teams use without knowing where it came from.
Nico van Zyl works as a freelance design engineer, mostly on design systems. Same shape as what I do: one senior person, working in both Figma and code.
Gabriel Valdivia works as a fractional design partner with early stage teams, embedding a few days a week rather than taking discrete projects.
Nick McEvily takes a similar fractional approach and prices it openly against the cost of the roles it replaces, which is a more honest framing than most.
In each case you are hiring a person rather than an organization. The senior person does the work because there is nobody to hand it to. That is the entire value and the entire risk, and any independent worth your money will say both halves out loud.
Signals that a shop is better than its website
Portfolios are a weak instrument. Every agency shows final screens, and final screens all look competent, because by the time something is finished the mistakes are invisible. So look for signals the marketing did not intend.
They show work that did not ship. A shop that only shows launched products is either very lucky or is hiding the projects that got cancelled, and every design career has those. The ones who show the abandoned concept and explain why it was abandoned are telling you they think in terms of decisions rather than deliverables.
They can describe a moment they were wrong. Ask about the last time user research contradicted their instinct. A real answer has a specific person in it, doing something the designer did not expect. A weak answer is a paraphrase of a process diagram.
Their case studies name a metric or admit there was not one. Both are fine. What is not fine is a vague claim of improvement with nothing behind it. Honest shops say "we did not have analytics on this, here is what we observed instead," and that sentence is more trustworthy than a rounded percentage.
They ask you hard questions on the first call. A shop that spends the whole first call presenting is selling. A shop that spends it asking who your users are and what happens if you do nothing is diagnosing. You want the second one, even if the first one is more flattering.
Red flags worth naming
A portfolio of only marketing sites. If every project is a landing page and you are building an application, the craft does not transfer. Product design is mostly about the ninth screen, and marketing design never has one.
Process as the pitch. When an agency leads with its methodology rather than its judgment, it is usually because the methodology is the thing that scales and the judgment is not evenly distributed across the team.
No opinion about your product. If you describe what you are building and the response is enthusiasm rather than a question, you have learned something. Good designers cannot help themselves. They start poking at it immediately, because that is what the job is.
The question that sorts everyone
Ask this: what did you decide not to build on your last project, and why?
A good designer will have an immediate answer, usually with some tension in it. A feature the client wanted that would have made the product worse. A screen they killed. A scope cut they had to argue for.
A weak one will not be able to name a single thing, because they built whatever they were handed. That is the whole answer, and it takes thirty seconds to get.
If you might not need an agency at all
A lot of teams reach for an agency when the real problem is narrow. One flow that loses people. One screen nobody can use. A design system that drifted into four button styles and three tab components.
Those problems do not need a firm and a statement of work. They need someone to look carefully and tell you precisely what is broken and in what order to fix it.
That is what a UX audit is. Fixed price, about a week, and you get a prioritized report and a fix first backlog your engineers can build from. If the answer turns out to be bigger than an audit can solve, you will at least know what you are buying before you buy it, which is more than most agency proposals give you.
If you want that run on your product, book a call and we can scope it in twenty minutes.