ConversionRetentionGrowthSatisfactionEase of useis where good design shows up.
I lead design for the Commerce domain at Cimpress, with end-to-end ownership of our B2B order-management platform. I lead a design team of 4 and guide a design group of ~15.
Open to Design Director and Head of Design roles · [Location / remote]
~$1M
saved a year on refunds
Recare · [measured or projected]
−48%
handling time on “Where is my order?”
WISMO · plus +70% CSAT
$500–600K
estimated annual saving in care call costs
Four platforms merged into one · [ballpark, confirm with real inputs]
4
senior designers hired
Into a team of 4, guiding ~15
~15
designers now prototype with AI
[confirm wording]
Acme Platform · 5 years · ~10,000 agents · my team's work, [your specific role]
Four platforms merged into one in a year, saving an estimated $500–600K annually in care call costs.
Every core function in one place, with the right information surfaced for care and sales agents. Quotes convert, tickets close and orders get fixed in about half the time.
Why four became one
The problem
To look up a single order or fix a problem, agents had to bounce between three other platforms.
How we got there
Over one year, including planning and aligning the teams, we pulled their information onto one central platform, moved the older platforms into keep-the-lights-on mode, and gradually sunset them.
What it changed
Agents and other teams now work in one place, which brought handling time down and CSAT up. An estimated $500–600K a year in care call costs [ballpark, confirm with real inputs]
How I led it
The mandate
[Who gave you the platform, and what was at stake for the business]
The bet
Treat four separate tools as one journey: know the customer, sell, track the order, fix what went wrong. [Why this was a bet, and who had to be convinced]
How we aligned
[How you keep product, engineering and the business on one roadmap across all four stages]
The outcome
Four platforms merged into one in a year. Separately, ~$1M a year saved on refunds through Recare. Handling time down 48% on order tracking and 37% on customer lookup, with CSAT up 70% and 40%.
What I'd do differently
[One honest lesson from the 5 years]
Before · After ·
The live prototype works best on a laptop.
Working prototype with made-up data. “Acme” stands in for the real Cimpress platform, which is under NDA.
~$1M
saved per year[Measured or projected, and over what period]
The problem
One major business was losing close to $1M a year on incorrect refunds and credits. Agents couldn't see what had already been paid out at the moment they issued a refund.
How we got there
A ladder of safeguards, each firmer than the last: refund and credit history in the workflow, a clear call-out for overpayment, then a block on further refunds once an order was fully compensated.
Trade-offs
We added friction, but only where it was useful. For overpaid orders we stopped trusting a warning to be read and built the rule into the product, which takes that decision away from the agent.
−48%
handling time[Baseline → result, and how it was measured]
+70%
CSAT[Baseline → result, and how it was measured]
The problem
“Where is my order?” [How agents and investigation teams answered it before, and where the facts disagreed]
How we got there
One order-details screen shared by care agents and investigation teams, so both work from the same facts. [Key decisions]
Trade-offs
[What each team gave up by sharing one screen]
−37%
handling time[Baseline → result, and how it was measured]
+40%
CSAT[Baseline → result, and how it was measured]
The problem
Agents had a lookup tool they searched and left. [What that cost agents and customers on each contact]
How we got there
From creating a customer to finding them again, we built one picture of who they are. [Research, key decisions and how it was rolled out]
Trade-offs
[What you chose not to show or build, and why]
[X]%
conversion[Baseline → result, and how it was measured]
−[X]%
time to order[Baseline → result, and how it was measured]
The problem
[Why building a cart and placing an order were getting in the way of selling]
How we got there
From building the cart to placing the order, we treated it as one sales workflow rather than a cart redesign. [Key decisions]
I run a team of 4 and guide ~15 designers across Cimpress Commerce, working directly with the people who own the business.
Product directorI report to
General managerof the domain, I report to
→
Me · Design LeadTeam of 4 · design group of ~15
↔
PartnersEngineering leads and managers, product managers and directors, business and client stakeholders
Operating model
Prototypes, not specs. The team builds quick prototypes in Claude from our design system, which expose gaps in requirements before anything is built.
Scale
The min–max framework sets the minimum every business gets and the maximum it can customise. [Example: a business it helped onboard]
Talent
I hired 4 senior designers. I look for curious people who take ownership and challenge assumptions. [How you grow them]
Rituals
A bi-weekly UX showcase, while there's still time to change course. [Critiques, design reviews, planning with product and engineering]
What I believe
Bring me the messy version
I'd rather see rough thinking early than polished work too late.
Make it tangible
Prototype, test, learn. I use AI and code to make ideas real before we overthink them.
Think beyond the screen
Look for patterns, systems and solutions that scale across businesses.
Day to day: find the real problem, draft UI within the week, stay with engineering until launch.
Operating model
Prototypes, not specs.
A strong design system lets the team turn an idea into a working prototype in a day, so gaps show up before engineering starts. [Confirm wording]
Before
Requirements→Specs→Build→Gaps found late
Missing cases surfaced during development, when they're slow and costly to fix.
Now
Requirements→Prototype in Claude→Gaps found early→Straight to developers
A shared skill and a coded version of the design system keep every prototype on-brand. [One example: a gap a prototype caught, and what it saved]
Side project · Stock market app · 2020 redesign, rebuilt in 2026
The same approach, on my own work.
I rebuilt a 2020 mobile redesign as a working app [confirm: with Claude], with real NSE closing prices and practice trades. It took the ideas from slides to something you can use.