When I joined LCCC, the billing process was both critical and fragile. Operators reconciled millions in renewable‑energy payments using PDFs that captured data at a single moment, quickly becoming outdated as figures changed. A single typo could invalidate an entire statement before anyone had opened it. The legacy workflow caused delays and mistrust in financial accuracy, and no one felt safe clicking “Approve.”, the billing process was both critical and fragile. Operators reconciled millions in renewable-energy payments using PDFs that captured data at a single moment, quickly becoming outdated as figures changed. A single typo could invalidate an entire statement before anyone had opened it.
Task
Redesign a billing flow used to reconcile national energy contracts — improving accuracy, transparency, and user trust in financial operations.
These statements underpin the UK’s Net Zero 2050 mission. The legacy workflow caused delays and mistrust in financial accuracy, and no one felt safe clicking “Approve.”
Design lens: Our early discovery mirrored the Discover → Define stages of the Double Diamond model, exposing hidden service gaps that shaped later design priorities.
As the sole designer across three Agile teams, I led UX for Scheme Operations workflows, partnering daily with our Product Owner and Lead Developer as a triad.
My goal was to translate complex financial logic into clarity and confidence for operators under pressure.
To achieve this I needed to understand the users for this problem I was designing for and re-immerse myself in their context.
Introducing Personas from our UX‑baselining project guided decisions throughout. They helped the team see operators not as job titles but as people balancing accuracy and accountability. Each persona anchored design discussions and directly informed what we visualised in the interface.
systems now happens in one flow.
internal validation tests.
with a single end-to-end flow.
statements after redesign.
through redesigned interface.
Early workshops felt like archaeology as we uncovered Slack threads and Excel trackers that kept the old process alive. Four “How Might We” questions became our compass, prevent context-switching, surface actions early, clarify statuses and make feedback visible and trustworthy.
That quote defined our north star: immediate clarity on what needs attention.
Overview
Problem Definition
HMW Statements
Ideation
Internal & Peer review
We moved from sketches to a prototype that unified expandable rows, inline evidence and a two-stage approval flow for audit compliance. Language became the infrastructure of trust. “Ready for Billing” became “Awaiting Review,” and “Approve Statements” became “Mark as Reviewed.”
To anchor users within a single view, we added a context pane that surfaced key data based on the user’s current job. Contract metadata, adjustments and policy hints appeared in one place so operators could drill deeper only when needed.
The experience evolved in line with the user’s context. As tasks changed, the interface adjusted what it showed and how it spoke to users. Messages reassured when risk was high and clarified intent when decisions mattered. This reflects Don Norman’s seven stages of action — particularly the execution-evaluation loop, where feedback and language close the gap between intention and outcome.
Testing focused on validating the concept and language changes. Feedback highlighted the importance of clear affordances and safe confirmation steps.
We responded using Jakob Nielsen’s heuristics for user interface design — especially Error Prevention and Visibility of System Status. Within a week we iterated the Reject option with reason field, confirmation pop-ups and status chips that visibly tracked progress.
Early workshops felt like archaeology as we uncovered Slack threads and Excel trackers that kept the old process alive. Four “How Might We” questions became our compass, prevent context-switching, surface actions early, clarify statuses and make feedback visible and trustworthy.
That quote defined our north star: immediate clarity on what needs attention.
This project reaffirmed that language is infrastructure — a real-world expression of emotional design (Don Norman). Changing two words achieved what no database migration could: it removed fear.
It also proved that context is reassurance. When information appears where it is needed and feedback closes the loop, users trust their own actions again. Trust here follows the logic of the Fogg Behavior Model: reducing friction in motivation and ability to produce reliable action.