BlueSG
BlueSG had no consistent way to bring customer evidence into product decisions. I started with one measurable service problem, then used that result to establish research, critique and design-system practices that teams could repeat without depending on one designer.
A short project brief before the detailed story: the setting, my responsibility, the people involved and the outcome we were working towards.
Product decisions varied by team, customer feedback was not collected reliably and UX outcomes were absent from leadership reviews. Eight in ten feature requests had no user validation, while 35% of development time went into avoidable fixes.
One successful project created permission to change the way the organisation worked. The routines made that change last.
Behavioural, operational and qualitative evidence did different jobs. Together, they helped define what the team needed to solve before choosing a direction.
Churn research and support data showed that unresolved service issues were affecting trust as much as interface inconsistency.
The first proof point needed to improve a real service outcome, not merely restyle the app.
Teams repeatedly rebuilt patterns and spent 35% of development time fixing avoidable experience issues.
Reusable components, design tokens and earlier critique could reduce downstream waste.
Most feature requests arrived without user validation and UX outcomes were absent from leadership reviews.
Research gates and outcome metrics had to become part of prioritisation—not optional design activities.
The NN/g maturity model provided a way to assess behaviour, process and adoption; it was not an external certification. The stages below separate completed change from the work that is still underway.
The Customer Relations Centre gave us a visible customer problem and an operational measure. Churn research, interviews and a focused service-design intervention made the value of UX tangible before I asked teams to change their wider practice.
Turnaround fell from 1.5 hours to 7 minutes; agent no-reply tickets fell 92%.
The CRC post-mortem showed management how research, collaboration and iteration led to the result. From there, we clarified cross-team responsibilities, set a shared voice and tone, and secured support for a broader research and design practice.
Design moved from a cosmetic service to an evidence-led product capability.
I introduced a practical Double Diamond SOP, discovery and design critiques, a shared research repository, an always-on participant panel, accessibility standards and a tokenised design system connected to delivery work.
Research became reusable, UX metrics reached leadership, and tokenised components reduced development cost by 25%.
The unfinished part is broader adoption: using customer evidence in more roadmap and operational decisions, helping other functions frame problems in the same way, and measuring whether the practices are actually used.
Stage 5 is a direction of travel, not a completed maturity claim.
The artefacts mattered when they helped the team choose. These are the questions, evidence and decisions behind the final direction.
The contact-centre journey had visible customer pain and measurable operational performance.
Use a focused service-design win to demonstrate value before proposing organisation-wide change.
Documentation alone would not change daily product and engineering behaviour.
Pair the Double Diamond SOP with critiques, workshops, research participation and delivery gates.
Inconsistent UI, accessibility gaps and repeated build work came from fragmented ownership.
Create shared tokens, components and standards while spreading decision ownership across functions.
The operating service and the interface formed one system. Each lens addressed a different part of the same problem.
Shaping the ecosystem, hand-offs and operating model.
Turning service decisions into clear, usable product behaviour.
The first intervention focused on the contact-centre service, where customer pain and operational performance were already visible. Turnaround fell from 1.5 hours to 7 minutes and agent no-reply tickets dropped 92%.
Post-mortems and stakeholder workshops connected the result to research, collaboration and iteration. We then established a shared voice and tone and a clearer cross-team cadence.
A Double Diamond SOP, weekly critique, research repository, participant panel and tokenised design system made the practice easier to repeat during live delivery work.
Customer evidence began to influence prioritisation beyond individual projects. Moving that responsibility further beyond Product and Design is the next stage, not a finished claim.
These constraints affected the direction, the order of work and the compromises made with the wider team.
A large transformation proposal would have sounded theoretical before teams had seen design improve a real outcome.
I began with a measurable CRC problem, shared the result and used that evidence to open a wider conversation about practice change.
Teams still had commitments to ship while new research and critique habits were being introduced.
We started with lightweight checkpoints at discovery and hand-off, then added rigour as the routines became familiar.
A wholesale redesign would have increased cost and delivery risk.
The design system grew around active roadmap work, allowing reuse and accessibility improvements to accumulate over time.
Stage 4 is the achieved state and Stage 5 remains in progress. The NN/g model was used as a diagnostic framework, not as an external certification.
Customer evidence now informs roadmap prioritisation.
Design and Engineering share a tokenised source of truth.
Critique, research and learning routines continue beyond individual projects.
UX outcomes are discussed with product leadership on a regular cadence.
My main contribution was connecting an early service result to a repeatable product practice and the infrastructure needed to sustain it.
Culture shifted faster when customer and business outcomes were visible, then reinforced through small routines that fitted the working week.
I would baseline adoption earlier and track research coverage, component reuse, accessibility and clear ownership of experience outcomes.