Two propositions came out of it: a design language that changes how the bank communicates with neurodiverse customers, and Nurominder, a credit score tool that prompts an action in time and carries the reason with it. Both were presented to Barclays' Chief Design Officer and leadership with a plan for rollout.



Sept – Dec 2022 · 3 Months
One of the UK's largest banks, working under inclusivity norms that were tightening across British banking. The brief was a digital proposition built for its neurodiverse customers rather than adapted to them afterwards.
Service Designer on a four-person Royal College of Art team, working directly with Barclays stakeholders throughout.
Service Design, UX Design, Design Strategy, Ethnographic Research, Workshop Facilitation, Insight-Opportunity Mapping, Prototyping, Client Management, Project Management
We ran four rounds, moving from the literature to the people it describes: what the spectrum actually covers, what Barclays and its regulator required, what 15 neurodiverse people said about managing money, and what six experts made of our reading of it.
Neurodiversity is a spectrum rather than a single condition, so the first job was to learn its range before assuming anything about it. Books, medical journal studies, a PESTLE analysis and competitor landscaping gave us two things: the breadth of need a proposition would have to cover, and the white space in the market where Barclays had room to move.

Barclays' design leadership set out what the bank wanted to be known for, and its VP of External Regulatory Engagement set out the legislation now shaping inclusive customer experience. Between them they drew the boundary the work had to sit inside, so the direction we took was aligned and compliant from the start.

We ran sessions with 15 neurodiverse people so that the pain points came to us first-hand rather than through the literature. Their accounts became personas and empathy maps, which is what kept every later decision anchored to a person and a moment rather than to a diagnosis.

We formed a panel of six experts across neurodiversity and inclusive design, and used it to test the direction rather than the artefacts. Their job was to catch where a well-meant design would be read differently by the people it was for, which is the difference between being inclusive and looking it.

Affinity mapping found the patterns, and sentiment and theme mapping then found where the feeling sat. That is what turned 15 accounts into the three barriers below.


None of the three barriers is about money. Each one is about how financial information arrives, and each feeds the next, so a credit score ends up recording a design failure rather than a financial one.
Financial information arrives dense and undifferentiated, written for one way of processing it. Where someone is sensitive to sensory load, decoding the statement is work in itself, and it has to be done before any decision about the money can begin.
The delay that load causes turns into something harsher, because people stop trusting their own judgement. Timings are missed, the score takes the damage, the lower score confirms the doubt, and the doubt makes the next action later still.
How a credit score actually works is never explained. Customers are asked to look after a number whose rules they have not been shown, so improving it becomes guesswork.
Two directions came out of it, one about the form information takes and one about who holds the decision, and everything designed afterwards answered to one of them.
Standard models of delivery assume one kind of reader. The goal was to make the form information takes something the customer chooses rather than something the bank fixes: offer choice, the inclusive design principle, rather than an accessibility toggle added afterwards. It also leaves the default calmer for everyone.
Credit management tools were the obvious deliverable and the smaller half of the job. The change worth making is in who holds the decision, because acting on time, and knowing why, rebuilds exactly the confidence the distrust loop takes away.
Five steps took us from opening the problem up speculatively to putting a rollout plan in front of Barclays. The expert panel and the participants came back in wherever a decision could go wrong.
We opened speculatively. What-If scenarios and Futures Wheel exercises pushed each idea out through its second and third order consequences, which asks what inclusive banking could become rather than what could ship next. Working back from those futures is what surfaced the immediate moves.

We ran Crazy Eights, SCAMPER and 10x10 against three targets taken straight from the barriers: reduce the sensory load, improve how information is processed, and improve the score itself. Volume was the point at this stage, and the concepts ran from a story box to a voice assistant, so the direction could be chosen from a spread rather than defended from the first idea.

We took the concepts back to the neurodiverse panel and the experts, who endorsed the focus and then corrected the detail. Two preferences came back consistently: familiar voice tones, and colour always paired with a sign rather than carrying meaning on its own, which is WCAG 1.4.1 applied to a language rather than to a screen. The same sessions settled the product question: what was needed was proactive credit score management.

Two propositions came out of the prototyping, and both were validated afterwards. The first is a design language built on semiotics, so that meaning is carried by signs a person already recognises rather than by financial vocabulary they have to hold in memory. That is recognition over recall, applied across a bank's communication.
The second is Nurominder, the credit score tool that speaks it. It sends a timely prompt and carries the reason for it alongside, so the knowledge arrives at the moment it can be used rather than separately from it.

We presented to Barclays' senior stakeholders with a route in rather than a concept alone: how the design language folds into the existing Barclays guidelines rather than sitting beside them, and what building Nurominder would take, either in-house or as a start-up in partnership with the bank. Business plans for rollout came with it.

These are four of the concepts the sessions produced, kept deliberately wide across physical and digital.




The work was handed over as journey maps, service blueprints, a product specification, a business model canvas and a business plan, covering both the experience and the case for building it.

A credit proposition neurodiverse customers can act on, and financial understanding that widens.
A route into cross-selling and customer base expansion across the broader UK neurodiverse audience.
A leading position in neuro-inclusive banking, and a standard the rest of the industry has to answer.
Rebuild the research method around the participant before you ask them anything about the product. The standard kit is itself a neurotypical instrument, and it filters what comes back.
Build the language first, then the products that speak it. Nurominder only works because a design language sits underneath it, and a product can be inclusive while the system around it is not.
Design the smallest timely intervention, and let it nudge rather than instruct. Within one spectrum, what relieves the load for one person adds to it for another, so any comprehensive tool will fail somebody.
Check Out Other Projects
Transforming Enterprise Data Protection
Led design pods across a two-year retainer, taking four self-initiated concepts into a global enterprise platform.
READ CASE STUDYRebuilding the UPI App India Was Given
I led the revamp of India's own UPI app, setting the four principles the product now runs on and shipping Split Expenses, Spends Analysis, and Family Mode.
READ CASE STUDYTransforming Credit-Card CX for 17 Million Customers
Designed and delivered a unified credit-card ecosystem, cutting support queries by 28%.
READ CASE STUDY