TLDR
Strategic Indices clients were testing J.P. Morgan’s indices in third-party tools, off the platform. I led design on an in-house backtesting workspace, discovery to tested design in eight weeks, held to four core actions. It shipped at the end of 2022.
01
The problem
Clients were testing J.P. Morgan’s indices somewhere else.
Backtesting shows how a strategy would have performed historically. Strategic Indices clients did it in third-party tools they found hard to navigate, so the work that decided an index happened off the platform. The brief brought it in-house: create, analyse and iterate without leaving. Two designers, four developers, a product owner, eight weeks. Every stakeholder had a good reason to show more.
02
Where the value was
A backtest is where an index sale starts, so I mapped each step against what it was worth.
A strategy gets tested against a client’s needs, talked through, then bought or structured as an index. The journey map put each step’s user goal beside its business opportunity, rated low to high. Choosing strategies rated high, because it showed what clients were interested in. So did adjusting an index directly in the platform.
03
The scope
Four core actions, held against every addition.
View strategies, test combinations, compare results, create benchmarks. The flow mapped the rerun, copy and edit loops before anyone prototyped them. A soft launch to selected clients, agreed to protect confidence in Strategic Indices as a whole, meant few early users to learn from, so the interface had to work on first contact. Design Studio sessions got design, engineering and product agreeing why before wireframes.
The decision
Traders don’t need to see the engine’s full complexity to use its full capability. Nothing on screen that isn’t doing work.
04
The result
Shipped at the end of 2022, with no baseline to measure it against.
- 8weeks, 0 to 1
- 4core actions
- 7people
Existing Strategic Indices users meant testing without outside recruitment. The capability shipped after the eight weeks, and little testing followed. The measure that mattered was time from idea to trade, and nobody had timed it in the old tools. I now treat a baseline as part of the work.
