GIA Platform

The team was busy. Design pass after design pass came through — review, redraw, re-litigate. Yet the platform wasn’t moving. Every sprint repeated the same conversation because no one had agreed on why any of it was being built.

I cleared the rituals. Three steps survived — and they actually moved the team.

I just needed three things to make it move.

Each step feeds the next, and the chain is visible. Once PMs, engineering, and business stakeholders could see how an Outcome became a flow and a flow became a Blueprint, the “why are you doing this?” question stopped landing in design reviews.

01 · Identify the Outcome — the Genie question

Removing the build constraint surfaces the north star and the real blocker in the same sentence.

02 · User story map — turn the outcome into a flow

The outcome turned into a flow — who does what, in what order, against which data.

03 · Low-fi wireframe — turn the flow into a blueprint

The flow turned into a blueprint — layout, state, and data contracts engineering builds to.

SOLUTION

Ask and validate what the visualization needs to answer.

2A · Forecasted Demand — the conflict

❓ QUESTIONS TO ANSWER

• Who has the biggest current value?
• Who will have the biggest future value?
• How much growth — or shrink — in between?

EXISTING VISUAL

⚠️ PROBLEMS

• X-axis = growth rate, Y-axis = future value, bubble size = current value — with no scale legend. Current values cannot be compared.
• No way to tell at a glance whether a value grew dramatically or shrank from a larger base.
• Drawn manually in PowerPoint — when bubbles overlapped, the deck builder repositioned them by hand. The visual stopped matching the data.
• Closer to a painting than a chart. Cannot scale.

✅ SOLVED BY

Setting the goal first and verifying the new visual answers it — which is what overcame emotional attachment to the existing version.

2B · Financial / Operational Metrics — the conflict

❓ QUESTIONS TO ANSWER

• Where does my account sit relative to its peers?
• Strong or weak — and on which metrics?
• Outlier, majority, or tail of the distribution?

EXISTING VISUAL

⚠️ PROBLEMS

• Distribution invisible.
• Outlier status invisible.
• Directional "strong / weak" read absent.
• Existing users coped on muscle memory; the next cohort wouldn't.

✅ SOLVED BY

Working backwards from the questions, listing the data needed, and picking a visual primitive — quartile box-plot cards — that made the answers readable at a glance.

Used as acting hands, offshore burned every overlap window on context transfer — not work.

Onshore / offshore isn’t a special model — everyone runs it. The failure mode is handing offshore acting work without the why; context breaks at the handoff, and the team works long but moves slow.

I split the designers into four profiles and approached each differently — the goal at every desk was autonomy and ownership, not handouts.

Re-evaluating talent instead of re-evaluating cost. Offshore became a velocity-and-talent lever, not a headcount one.