Disputes · Payments · Risk · Process design

Making the opaque parts of payments interesting.

I am a communication design graduate working as a chargeback analyst. I build tools and I am interested about payment rules, dispute logic and risk. Every project here is made with AI, and these help me navigate disputes logic and decision making.

CurrentlyChargeback Analyst, Monzo
BasedEEA / UK
FocusFintech disputes, card payments, threat mitigation and AI risk
Open toFraud strategy, payments analytics
5Years across disputes, AML and trust & safety operations
4Interactive case studies, designed and built end to end
3Institutions: Monzo, NatWest Group, Accenture
12+Dispute categories mapped into working decision logic
How I work

Process over perfection.

A finished product is the tip of the iceberg. Underneath sit research, dead ends, reframes and iteration. This is the shape of how I think.

01

Map the problem

Before any wireframe, I dissect the regulatory, operational and human constraints. What does the user need to decide, and what does compliance require?

02

Model the flow

Complexity becomes decision trees, state diagrams and logic maps. The goal is to make the invisible system visible, first to myself, then to the user.

03

Build in the open

Rough prototypes beat perfect mockups. I test logic with real data, real edge cases and real confusion. Iteration is the design.

04

Ship and measure

Launch is a milestone, not an endpoint. I watch where users stall, where rules break, and where the next version begins.

Selected work

Navigating payment disputes.

Each piece is live and interactive. Open any of them and it runs as an overlay on this page, rule logic and sources intact.

01
Live Interactive tool · Mastercard UK · Issuer side

Chargeback Triage Advisor

A structured triage tool for UK issuer dispute analysts. Enter the case details and it returns a filing recommendation, an evidence checklist and an indicative win probability, grounded in Mastercard reason codes and UK filing windows.

Problem

Analysts triage high dispute volumes under time pressure, with the rules spread across dense network guides.

Approach

An eight branch decision tree, progressive disclosure, and a checklist generated for each route.

Outcome

A working triage tool, used as a portfolio demo and as a training reference for new analysts.

Runs as an overlay
DISPUTE 4837 · FRAUD 4855 · NOT PROVIDED 4860 · CREDIT FIGHT DECLINE FIGHT CAUTION FIGHT ACCEPT REASON CODE RECOMMENDATION
Fig. 01 · Decision logic8 branches
02
Live Simulator · Crypto merchants · EEA / UK

European Crypto Card Chargeback Triage Simulator

Card disputes at crypto merchants fail in unfamiliar ways. This simulator models SCA liability, volatility driven friendly fraud and on chain delivery signals, then explains its reasoning rule by rule.

Problem

Friendly fraud spikes when asset prices drop. Cardholders dispute purchases they made, minutes after a crash.

Approach

A scoring engine across tenure, velocity, volatility and custody signals, with the full rule logic visible.

Outcome

A transparent simulator with demo cases, exportable results and a printable view.

Runs as an overlay
72 / 100 WIN PROBABILITY 0 100
Fig. 02 · Risk gaugeScore 72 / 100
03
Live Longform · Payments strategy · 6 min read

The True Cost of a Chargeback

A £65 dispute is never just £65. A longform breakdown of the dispute economy: the hidden cost stack on both sides of the network, the friendly fraud measurement problem, and what good actually looks like. Includes a cost model you can adjust yourself.

Problem

Stakeholders underestimated the true cost of a dispute by about sixty percent.

Approach

Scrollytelling structure, published sources cited inline, and an interactive model with deflection built in.

Outcome

A six minute read that reframes the question from cost per case to cost per dispute ecosystem.

Runs as an overlay
FACE VALUE £65 NETWORK FEE £20 GOODS LOST £29 STAFF TIME £17 TOOLING £11 £65 in, £143 out. ≈ 2.2× FACE VALUE
Fig. 03 · The cost stack£65 → £143
04
Live Explorer · P2P markets · Two rail problem

P2P Crypto Disputes: Where the Loss Lands

Peer to peer crypto trades settle on two rails that never touch. Pick a failure point, from escrow disputes to fake payment proofs, and see who bears the loss and which rulebook actually applies.

Problem

When a P2P trade goes wrong, users cannot tell who bears the loss, or which rules even apply.

Approach

A five step lifecycle map, four failure scenarios and a mini triage tool for your own case.

Outcome

An explorer that answers the only question that matters: if this goes wrong, who pays.

Runs as an overlay
FIAT RAIL CRYPTO RAIL REVERSAL 1 · ESCROW 2 · PAYMENT 3 · CONFIRM 4 · RELEASE 5 · AFTERMATH BANK ON CHAIN
Fig. 04 · Two rails, one break5 stages
Background

Teams where I have worked.

2023 · Present
Chargeback Analyst
Monzo Bank

Managing high volume disputes for a leading digital bank. Focused on minimising revenue leakage through accurate reason code interpretation and evidence retrieval.

Dispute resolution · Mastercard rules · Evidence gathering
2022 · 2023
Operations Analyst
NatWest Group

Built strong relationships with internal and external stakeholders to streamline AML processes and ensure regulatory alignment.

AML · Process improvement · Stakeholder management
2021 · 2022
Trust & Safety Analyst
Accenture Portugal

Handled sensitive digital safety cases, ensuring platform compliance and user protection. Developed strategies to mitigate fraud risk in high traffic environments.

Trust & safety · Fraud mitigation · Platform compliance
Education
Postgraduate
MSc Management with Marketing
Master level

Applied knowledge of customer engagement, strategies and management. Passed with Merit.

Undergraduate
BSc Communication Design
Bachelor level

Visual systems, typography and process design.