When the composer rewrites the score, the orchestra does not get to keep playing the old one. It also does not get to stop playing while it learns the new one.
Here is the uncomfortable truth: most payment firms will treat PSD3 and the Payment Services Regulation as a compliance project. That is how they will lose money on it.
The legal change is real. Europe is replacing one directive, PSD2, with two instruments. PSD3, a directive, governs licensing and supervision and still needs to be written into each country's law. The PSR, a regulation, carries the conduct rules (authentication, fraud, open banking, transparency) and applies directly and identically across every member state. The aim is to end the patchwork of national interpretations that PSD2 produced.
But the real change is commercial. For the first time, a large share of fraud losses moves firmly onto the provider's balance sheet. A customer tricked by someone pretending to be their payment provider will, under the agreed text, be entitled to a full refund once they report it to the police and to the provider. A provider that fails to run adequate fraud prevention carries the loss. That is not a policy line. It is a P&L line.
And the clock is shorter than it looks. As of this writing, the texts are agreed but not yet published in the Official Journal; industry trackers expect publication around the end of 2026. From entry into force, most rules apply after 21 months, and the payee-name check after 27 months. Existing payment institutions have up to 27 months to demonstrate they meet the new licensing requirements. Twenty-one months sounds like a long runway. Subtract a budget cycle, a vendor selection, an integration, a test period and a re-authorisation file, and it is not.
Here is how a Maestro would evaluate, plan and communicate this transition, movement by movement.
What is actually changing
Legal framework. PSD2 splits into PSD3 (licensing and supervision) and the PSR (conduct rules that apply directly in every member state). E-money institutions are folded into the payment institution regime. Why it matters: less room for national interpretation, and your authorisation must be re-evidenced.
Verification of payee. Providers must check that the payee's name matches the account identifier before a credit transfer, and warn the payer of a mismatch. Why it matters: a product, data and customer-experience change across every transfer flow.
Strong customer authentication. SCA rules are tightened and made more accessible, including for customers without smartphones. Why it matters: authentication journeys and exemptions need redesign, not tweaks.
Fraud liability and spoofing. Full refunds for bank-employee impersonation fraud once reported. Liability where fraud monitoring or payee checks fail. Online platforms liable to providers if they ignore notified fraudulent content. Why it matters: fraud becomes a direct cost and a negotiating point with partners.
Open banking. Banks may not place listed obstacles in the way of authorised third parties, and customers get a dashboard to see and revoke data access. Why it matters: an opportunity for fintechs, and a build and governance cost for account providers.
Access to cash. Retailers may offer small cash withdrawals without a purchase, and without needing a payment licence. Why it matters: relevant for merchant-facing and acquiring businesses.
Note the operative word: agreed. Until publication, the law in force is still PSD2. Plan against the agreed text, and keep a watch on the final version.
Tempo: read the real clock, not the headline one
Tempo is not about speed. It is about the right pace for each part of the work.
The mistake I see most often in regulatory change is a single tempo for everything: either a panicked sprint that exhausts the team before the hard part arrives, or a relaxed andante that wakes up in month eighteen.
A Maestro sets different tempos for different sections:
Slow and deliberate now: the gap analysis and the business decisions. These are the choices you cannot easily reverse.
Steady through the middle: build, integration and vendor work, run to a fixed weekly rhythm regardless of news.
Fast at the end: testing, customer communication and cut-over, rehearsed in advance so speed does not mean improvisation.
Then work backwards from the application date, not forwards from today. Fix your own internal deadline at month 15, not month 21. The six months you hold back are the buffer for the delay that will happen.
Silence: listen before you plan
Silence is the movement about hearing what is not being said.
Every firm will commission a gap analysis. Most will miss the most valuable input, because it does not sit in the regulation. It sits with the people closest to the fraud.
Before you build the plan, go and listen:
Your fraud operations team knows which scam patterns are rising, where your monitoring is blind, and how many of last year's losses would have become refunds under the new rules.
Your customer service team hears how customers describe being tricked. Those words are your future verification warnings.
Your complaints and disputes data shows where liability will land first.
Ask one question in each room: If the new liability rules had applied last year, what would it have cost us, and where? The silence after that question is where your real exposure lives.
Rehearsal: run the liability before it runs you
Rehearsal is where organizations become capable rather than merely informed.
Most firms will have a PSD3 plan on a slide. Very few will have played it. Schedule the rehearsal early, while the stakes are still hypothetical:
Run a spoofing simulation. A customer reports that someone posing as your staff talked them into a transfer. Walk it through fraud, customer service, legal, finance and communications in real time. Who decides on the refund? How fast? Who tells the customer? Who claims against the platform where the scam started?
Run a payee-mismatch scenario. A business customer's legitimate payment is flagged because the name does not match. Find out how much friction you create for good customers, not only how many bad payments you stop.
Find the seams. The failures will not be inside departments. They will be in the handoffs.
Fix the handoffs, then run it again.
Interpretation: the score is fixed, the performance is yours
Interpretation is the movement about what leaders do with the score they are given.
Two firms can read the same regulation and produce two very different businesses. One reads it as a cost to minimise. The other reads it as a change in the market to exploit.
This is where the evaluation happens. I would put four questions to the leadership team:
Exposure: What is our fraud liability under the new rules, measured on last year's data, and what is the trend?
Capability: Which of the six areas can we deliver ourselves, which need a partner, and which partners are themselves now liable?
Economics: What does compliance cost, and which products stop making money once fraud losses sit with us?
Opportunity: Where does the new framework open a door? Fewer open-banking obstacles, stronger payee checks as a trust feature, a cleaner licensing route across Europe.
Some products will not survive question three. Better to discover that in a board paper now than in a loss report later. And some firms will find in question four a positioning advantage their competitors are too busy to see.
Orchestra: one programme, not six workstreams
Orchestra is the movement about collective intelligence.
PSD3 and the PSR touch compliance, product, engineering, fraud, finance, legal, customer service and partnerships at once. Run as six separate workstreams, they will produce six separate truths.
A Maestro builds one transition office with three rules:
One view of exposure, owned by a named executive, updated monthly to the board.
One map of third parties. The agreed text extends liability to technical service providers and payment scheme operators in certain cases, and gives providers a claim against platforms that ignore fraud notices. Every contract with a processor, authentication vendor, scheme or marketing platform needs reviewing for who pays when things go wrong.
One voice to customers, partners and regulators. Mixed messages here cost trust and invite supervisory attention.
Ego: the dangerous sentence is "we're already compliant"
Ego is the movement about the leader's need to be right.
The most expensive sentence in this transition is "we're already compliant with PSD2, so this is an upgrade." It is not. The liability model has changed, and the authorisation must be re-evidenced.
The second most expensive move is delegating the whole thing to the compliance function. Compliance can tell you what the rules require. It cannot decide which products to reprice, which partners to renegotiate, or how much fraud risk the business is willing to carry. Those are leadership decisions, and the board should see them as such.
Ego also shows up in communication: the temptation to project certainty about dates and details that are not yet final. A leader who says "here is what we know, here is what we don't, here is what we are doing" will be trusted when the details land.
Conducting the communication: internal and external
A transition of this size is won or lost in how it is explained. The principle is the same for every audience: say what is known, what is not yet known, and what you are doing about it. Tell people early, tell them consistently, and never promise customer protections before the law and your systems actually deliver them.
Internal communication
Board. Message: exposure, cost, strategic options, decisions required, and risk appetite for fraud liability. When: now, then monthly. Channel: board paper and standing agenda item. Owner: CEO with CFO and CRO.
Executive team. Message: programme scope, owners, the internal deadline at month 15, and budget. When: now, then fortnightly. Channel: transition office review. Owner: CEO.
Fraud, compliance and engineering teams. Message: detailed requirements, dependencies and test plan. When: ongoing. Channel: programme cadence and documentation. Owner: programme lead.
Customer-facing teams. Message: what changes for customers and when, what to say, and how refunds and payee warnings will work. When: six months before go-live, refreshed before launch. Channel: briefings, scripts and FAQ. Owner: head of customer operations.
All staff. Message: why this matters, how it protects customers, and the fact that impersonation scams borrow staff identities. When: at programme launch and at go-live. Channel: all-hands and internal update. Owner: CEO.
External communication
Retail customers. Message: the new protections, payee-name checks, how to recognise impersonation, and what you will never ask them to do. When: two to three months before each change goes live. Channel: app messages, email and website. Owner: customer and marketing lead.
Business and merchant clients. Message: the effect of payee checks on their payments, trading-name requirements on statements, and any pricing or contract change. When: six months before go-live. Channel: account managers and written notice. Owner: commercial lead.
Partners and vendors. Message: contract and liability changes, technical requirements and timelines. When: within the first six months. Channel: formal letters and joint working sessions. Owner: partnerships and legal.
Regulators. Message: your transition plan and re-authorisation file. When: early, and well before deadlines. Channel: supervisory dialogue. Owner: head of compliance.
Investors and media. Message: cost, risk and opportunity, framed with discipline. When: with results or at a strategic milestone. Channel: results commentary and statements. Owner: CEO and CFO.
Two practical rules. First, do not use the new fraud rules as a fear campaign; customers remember who frightened them. Second, the customer message about impersonation fraud is also a staff message, because scammers borrow your employees' names and voices.
The transition plan at a glance
Prepare (now until publication). Listen to fraud and service teams, size the exposure, and take board decisions on scope and risk appetite. Movements: Silence, Interpretation.
Design (months 0–6 after entry into force). Gap analysis against the final text, vendor and contract review, re-authorisation plan and budget. Movements: Tempo, Orchestra.
Build (months 6–15). Payee checks, SCA redesign, fraud monitoring and the open-banking dashboard, with the internal deadline at month 15. Movements: Tempo, Orchestra.
Rehearse (months 12–18). Spoofing and payee-mismatch simulations, customer-facing scripts and partner tests. Movement: Rehearsal.
Perform (months 18–21). Customer communication, cut-over and monitoring. Movements: Tempo, Ego.
Sustain (months 21–27). Payee-check go-live, licence transition completed, lessons learned. All six movements.
FAQ
When do PSD3 and the PSR apply?
Not yet. The texts are agreed but, at the time of writing, still awaiting publication in the Official Journal, expected around the end of 2026. Most rules then apply 21 months after entry into force, with the verification-of-payee provisions applying after 27 months.
What is the difference between PSD3 and the PSR?
PSD3 is a directive covering licensing and supervision and must be transposed into national law. The PSR is a regulation covering conduct rules such as authentication, fraud prevention and open banking, and applies directly across the EU.
Who pays for impersonation fraud under the new rules?
Under the agreed text, where a fraudster poses as an employee of the customer's payment provider, the provider must refund the customer in full once the customer reports it to the police and to the provider. Online platforms that ignore notified fraudulent content can be liable to the provider.
Do existing payment institutions need a new licence?
Existing institutions will need to show they meet the new requirements. The agreed text allows them to continue operating for up to 27 months after entry into force while they provide that evidence to their regulator.
Is PSD3 only a compliance issue?
No. It changes fraud economics, partner liability and open-banking access. Leaders should treat it as a strategic decision about which products, partners and risks the business wants to carry.
Conclusion: play the new score, not the old one
Europe has rewritten the score for payments. The firms that come through this well will not be the ones with the thickest compliance binder. They will be the ones whose leaders listened to the people closest to the fraud, rehearsed the hard moments before they happened, interpreted the rules as a strategic choice, and told their people and customers the truth along the way.
The rules are the composer's. The performance is yours.
If you want to know which of the six movements your leadership team is weakest in, take the Conductor Score. To get a new movement every week, subscribe to Conducting Leadership.
Timelines and provisions reflect the final compromise texts published by the Council of the EU in April 2026 and industry commentary as of October 2026. The package has not yet been published in the Official Journal; This article is leadership commentary, not legal, regulatory or compliance advice.