36 PSPs selected for the digital euro pilot — what it means

On 14 July 2026, the ECB announced the payment service providers selected for the digital euro pilot: 36 PSPs, chosen from 57 applications, across 19 countries including Croatia. Banks and non-bank providers, selected for broad representation of business models, size and geography.

It is the most concrete thing that has happened to the digital euro in years. It is also being widely misread. Let us separate the two.

What was actually announced#

The facts, unembellished:

Selected	36 PSPs
Applied	57
Countries	19, including Croatia
Mix	Banks and non-bank providers
Announced	14 July 2026
Development starts	Q3 2026
Live pilot	H2 2027, running ~12 months

What it is not#

It is not a launch. The digital euro does not exist in any public form, and will not in 2027.

It is not a public beta. This is the one that matters most, because the word "pilot" invites the wrong picture. The live pilot beginning in H2 2027 uses a beta digital euro, and the users are ECB and national central bank staff. Not customers. Not the public. Not a segment of early adopters at the 36 institutions.

Nobody's customers are getting a digital euro in 2027.

It is not a commitment to issue. The regulation is still in the EU legislative process, and the decision on whether to issue the digital euro will only be considered once that process completes. The Eurosystem is building so it can be ready. Readiness is not a decision.

What it genuinely signals#

Now the part that matters.

The thing is being built#

Development starts in Q3 2026. Not designed, not consulted on — built. After years in which the digital euro was a set of documents and a legislative debate, there is now an implementation programme with participants and a date.

For anyone at a PSP who has been treating this as perpetually theoretical, that assumption just expired.

57 applications is the interesting number#

Everyone is reporting the 36. The number that tells you more is the 57.

Fifty-seven PSPs volunteered to do unpaid, early, difficult integration work against a draft specification for a currency that has not been legally authorised yet. That is not what an industry looks like when it thinks something will not happen.

Twenty-one of them were not selected — and they still have the same obligation coming.

Geographic spread is a design statement#

19 countries, including Croatia. The euro area has 20 members. This is not a pilot among the big five markets.

That matters because harmonisation is the entire point of the scheme. A rulebook exists so that a user gets the same experience regardless of which country they are in or which PSP they use. You cannot validate that claim by testing in France and Germany. Including small markets is how you find out whether the scheme actually works everywhere before it is too late to change.

Banks and non-bank providers#

The mix reflects the distribution model. The digital euro is distributed only through supervised PSPs — banks, payment institutions and e-money institutions — and the pilot's composition takes that seriously rather than defaulting to incumbents.

What it means if you are one of the 36#

Your build starts now, against rulebook v0.91, which is explicitly a draft with unfinished sections — including the functioning and enforcement of the holding limit.

You are not implementing a specification. You are helping find out where the specification is wrong. That is genuinely valuable, and it is a different job from the one your delivery plan probably describes. The feedback loop is real: v0.91 itself incorporates feedback from the market consultation that ran June to October 2025.

What it means if you are one of the other 21 — or the rest#

Here is the thing worth internalising: not being selected changes your timeline not at all.

If you are a credit institution, you will be obliged to offer basic digital euro services to customers on request once the digital euro launches. And basic services are free for individuals. That obligation does not care whether you were in the pilot.

What you lost by not being selected is learning time, not deadline. So take the learning where you can get it:

• The pilot documentation is public. Technical documentation is on the ECB website, including back-end API specifications in YAML. You can read the contract you will eventually implement.
• The RDG publishes progress reports. They show what is being worked on before it lands in a published draft.
• Do not start coding. v0.91 is a draft. Code written against it now is code you write twice.

What you should do in the next twelve months is decide — not build. Work out which role you need (distributing PSP, acquiring PSP, or both) and make the build-vs-buy call, because that decision has the longest lead time and gets more expensive the later you make it. Note that PSPs may outsource development and operations to a TSP, while retaining regulatory responsibility.

The honest summary#

The pilot announcement does not move your deadline. It removes your excuse.

The digital euro has been easy to file under "someday" because it had no implementation date. It now has 36 institutions building against real infrastructure starting Q3 2026, and a live pilot in H2 2027.

The gap between now and ~2029 looks generous. It is not, because of what sits in the middle: a window where every credit institution in the euro area needs an integration at once, drawing on the same small pool of people who have done this before. The institutions that decide how they are solving it in 2026 will be buying in a normal market. The ones that start in 2028 will be buying in a scramble.