Aastha Shree. ← Back to portfolio

Case study · Goldman Sachs

From 60 tools to
one platform.

Consolidating a fragmented analytics estate into a single product — and getting 1,100+ people in a regulated function to actually use it.

Role  VP, Senior Product Manager Scope  1,100+ users · APAC, EMEA, Americas Timeframe  2021 — present
60+ → 1overlapping tools consolidated into one suite
<5% → 20%divisional adoption — a 4× lift
~62Khours/yr of cost-avoidance by 2025
3regions on one standardised platform

The situation

When I took ownership of the division's analytics capability, the work was spread across 60+ overlapping tools. Teams had built whatever they needed, wherever they could — which meant duplicated effort, inconsistent numbers, and no single place anyone trusted. For a function that exists to give the firm assurance, the analytics underneath it were anything but assured.

The symptom looked like too many tools. The real issue was that no one owned the thing as a product.

Framing the real problem

It would have been easy to run this as a migration: pick a winning tool, move everyone, declare victory. I resisted that. Migrations move people off old things; they do not make anyone want the new one. Adoption had sat under 5% precisely because previous efforts solved for tooling, not for the user's reason to switch.

So I reframed it as a product problem with three parts: there was no single owner, no shared definition of the work, and no compelling reason for a busy auditor to change how they already got their job done. Fix those three, and consolidation would follow. Chase consolidation alone, and I would end up with a 61st tool nobody used.

“A platform people avoid is just a more expensive version of the mess you started with.”

Constraints I had to design around

The bets I made

01

Treat it as one product, not a cleanup

I set a product vision and OKRs for a single standardised suite, with a multi-quarter roadmap — so decisions were made against where we were going, not which tool shouted loudest.

02

Consolidate by decision, not by inventory

Rather than port 60+ tools, I decided deliberately what to keep, merge, and retire — anchored on the handful of workflows that actually drove the function's decisions.

03

Make adoption a first-class metric

Adoption was an OKR from day one, instrumented and visible — not a hope we checked at the end. If people were not using it, that was my problem to solve, in the product.

04

Build trust into the product, not around it

Governance and controls were designed in from the first requirement, which is what let a regulated function actually rely on the platform rather than route around it.

05

Earn the switch through enablement

I drove change with training, embedded workflows, and tight feedback loops — bringing analytics into the reviews people already ran, so the platform met them where they worked instead of demanding they come to it.

What shipped

Results

What I would do differently

Two things. First, I would have invested in change management and user segmentation earlier — sequencing the rollout by appetite, starting with the teams most ready to switch and letting their wins pull the hesitant ones, rather than treating all 1,100 users as one audience. Second, I would have been more aggressive about retiring tools sooner; leaving legacy options alive a little too long gave people a reason to hedge, which slowed the very adoption I was measuring.

My role

I owned the product end to end: vision, roadmap, prioritisation, and the adoption strategy. I partnered closely with engineering, data, and audit stakeholders, kept governance in the loop from the first requirement, and led two Product Managers through delivery.

Want the rest of the story?

Happy to walk through the decisions, the trade-offs, and the parts that did not go to plan.

Email Aastha ↗ Back to portfolio