Shipped work (30%). We look for published case studies with live links: what the problem was, what the agency built and what changed after launch. Work that is live in a client’s site or product counts for more than concepts, pitches or unlaunched redesigns.
How we rank design agencies
DesignAgencyIndex ranks design agencies on the work they have shipped. Every agency on the site gets a score out of 10, built from five weighted criteria and checked against the agency’s own website and live case studies. This page explains what we look at, what we leave out and how each list is put together.
The score
The five criteria make up the score in fixed shares, so you can see how much each one counts.
| Criterion | Weight | What it measures |
|---|---|---|
| Shipped work | 30% | Live, published work behind the agency’s claims |
| Client roster | 20% | Who the agency has worked with, and on what |
| Category depth | 20% | How much of its work sits in the service or industry of the list |
| Verified recognition | 15% | Awards and press we can link to |
| Recent output | 15% | Strong work launched in the last two years |
What we check
Client roster (20%). Named clients are weighed by the scale of the company and by repeat engagements. The work has to be visible in the client’s live site or product, not only as a logo on the agency’s homepage.
Category depth (20%). An agency ranks higher on a list when a real share of its portfolio sits in that service or industry. A studio with ten fintech projects outranks one that lists fintech as one line on a long services menu. This is why the same agency can hold different positions on different lists.
Verified recognition (15%). Awards and press count only when we can link to the source. Unlinked badges and self-reported awards are left out of the score.
Recent output (15%). We weight work launched in the last two years, so one famous project from a decade ago cannot carry the score on its own.
Where the information comes from
Facts about an agency come from its own website first: location, founding year, services, clients and case studies. When another source says something different, the agency’s own site wins. Case studies are linked directly, not through a portfolio homepage. Third-party sources are used for recognition only, such as award sites and design and industry press.
We do not use paid surveys, self-submitted forms or client review scores. Reviews are hard to verify and easy to game, and they say little about the quality of the work itself.
How the lists are built
Who qualifies. The agency is active, has a working website on its own domain and has published case studies in the category of the list.
Minimum size. A list goes live only when at least eight agencies qualify for it.
Order. Agencies are ordered by score. When two scores tie, the one with stronger recent output ranks higher.
Written for each list. Every entry says who the agency suits and when to look elsewhere, and that text is written for the specific list, not copied between pages.
What does not affect rankings
Payment. We do not sell placements or sponsored spots. No agency can pay to be listed or to move up.
Size and rates. Headcount and pricing are not part of the score. A small studio with strong work can outrank a large network.
Location, except on city lists, where an agency needs a team in that city.
Marketing claims that the agency’s published work does not back up.
Updates and corrections
Rankings are updated for 2026. We revise an entry when an agency launches significant new work, changes focus or closes, and we remove agencies whose websites go offline or redirect to another company.
If you spot a fact that is wrong or out of date, let us know with a link to the source on the agency’s website.