Case Studies

The work,
by the numbers.

Three full case studies from Gopuff and S'well — each showing the specific problem, how I approached it, and what shipped. No vanity metrics. Just outcomes that moved the business.

Gopuff · Marketplace & Delivery Ops · 2021–2022

Building the company's first
real-time order platform
from nothing.

Gopuff operated 600+ micro-fulfillment sites across the US and UK with no unified view of what was happening to an order after it left the shelf. Late deliveries were invisible until a customer complained.

80%

Reduction in P90 deliveries over 60 minutes

1wk

Time from kickoff to working MVP

600+

Sites on platform within one month

The Challenge

Gopuff had no real-time visibility into the order lifecycle after dispatch. Drivers, dispatchers, and operations managers were flying blind — they couldn't see where an order was, why it was late, or when it would arrive.

The result: a P90 late-delivery window (deliveries over 60 minutes) that was both large and invisible. The business couldn't fix what it couldn't see.

The Approach

I scoped an MVP deliberately narrow: one team, one site, one question — can we surface real-time order state in a way that actually changes dispatcher behavior? We went from kickoff to a working product in one week.

The key decision was treating the first pilot as a learning instrument, not a proof of concept. We defined explicit success criteria up front, ran tight feedback loops with dispatchers on site, and used what we learned to rebuild the data model before scaling.

Full productization shipped within the month. We then ran a structured rollout — one region at a time, with clear readiness criteria before each expansion — until the platform covered 600+ sites across the US and UK.

The Outcome

P90 late deliveries over 60 minutes dropped by 80%. Operations teams went from reactive firefighting to proactive intervention — they could see a delivery at risk and act before the customer knew.

The platform became the operational backbone for all subsequent delivery optimization work, including the Uber integration that reduced costs during low-supply periods.

"The hardest part wasn't building it — it was resisting the urge to build everything at once. The MVP worked because it answered one question clearly. Everything else came from that."

What I owned

Product strategy & scope — defined the MVP boundary, success criteria, and phased rollout plan

On-site pilot design — embedded with dispatchers to test and iterate before scaling

Technical partnership — collaborated with engineering on data model decisions that made scale possible

Nationwide rollout — led phased expansion across 600+ sites with regional readiness gates

0 → 1 Real-time Systems Operations Pilot Design Nationwide Rollout Delivery Ops

Gopuff · Driver Experience · 2022–2024

Rebuilding the driver app —
and saving $30MM+
in the process.

As Head of Product for Driver Experience, I owned the entire driver mobile app for Gopuff's gig workforce — the critical last mile of every single delivery. Low NPS, high churn, and a product that hadn't kept up with the scale of the business.

+15

Driver NPS points gained

$30MM+

Annual cost savings from Ops initiatives

25%

Reduction in order mispack rate

The Challenge

Gopuff's driver mobile app was a product built for a smaller, simpler version of the business. Driver NPS was low, onboarding took too long, and critical flows — pay, scheduling, pick-up, drop-off — had accumulated years of technical and UX debt.

At the same time, the operations organization needed the product team to drive cost savings at scale: inventory shrink, mispacks, and dispatch efficiency were all bleeding money.

The Approach

I ran two parallel workstreams. On the driver experience side, I led end-to-end redesigns of the highest-friction flows: a new pay model with clearer earnings visibility, a faster onboarding sequence, a go-on-break feature that drivers had been requesting for over a year, and rebuilt pick-up and drop-off flows.

Every change went through a rigorous user research and pilot cycle — in-app feedback, on-site observation, and A/B testing before any nationwide rollout. We shipped to one cohort, measured NPS delta, then scaled.

On the ops side, I managed a 20-person cross-functional team to deliver AutoDispatch (intelligent driver assignment), Chain of Custody (order tracking), and Dynamic Pricing — each designed to reduce waste and improve margin across the network.

The Outcome

Driver NPS increased by +15 points — a material shift in a workforce where satisfaction directly correlates with supply availability and delivery speed.

The ops initiatives delivered $30MM+ in annual cost savings, $2MM in reduced inventory shrink, and a 25% reduction in the order mispack rate. I also brokered the Uber integration — a strategic third-party partnership to cover low-supply periods — reducing delivery costs and ensuring service continuity.

"+15 NPS points doesn't sound like much until you realize that every point represents thousands of gig workers deciding whether to log in tomorrow. Driver supply is a product problem."

What I owned

Full driver mobile app — end-to-end ownership of all flows and the product roadmap

20-person XF team leadership — engineering, design, ops, and data across two product orgs

Uber partnership — scoped and launched third-party integration to solve supply gaps

Cost initiatives — AutoDispatch, Chain of Custody, Dynamic Pricing, anti-fraud programs

Mobile App Team Leadership NPS Gig Workforce Cost Reduction Strategic Partnerships

S'well · Innovation · 2018–2019

First PM hired. Fastest launch
in company history. One year.

S'well was a 100-person consumer goods startup with $100MM in revenue and zero product management discipline. I was the first PM they'd ever hired — brought in to build the function, launch a new category, and prove that process could coexist with speed.

$50MM+

Revenue from 22 launches in year one

100k+

Units sold — fastest launch in company history

100%

On-time, on-budget across all 22 launches

The Challenge

S'well had grown fast on the strength of a single product line — beautiful stainless steel water bottles. Now they wanted to expand into food storage. No product management process existed. No consumer testing platform. No structured roadmap. Just a mandate to move fast and not break the brand.

As the first PM, my job was to build that infrastructure while simultaneously launching the new category — a situation that is either exciting or terrifying, depending on your disposition.

The Approach

I started with the foundation: market sizing, competitive analysis, and 200+ consumer insights to understand the food storage jobs-to-be-done that S'well was positioned to own. That research defined the product roadmap, positioning, and user requirements for the new category.

In parallel, I built S'well's first consumer testing platform — negotiated supplier contracts, conducted cost-benefit analyses, and implemented the end-to-end system that let the team validate concepts before committing to production.

I also redesigned the product development process itself — a new cross-functional framework that aligned 13 partners up to SVP level and reduced product timelines by 30%. The CEO called it the most effective framework in company history.

The Outcome

We launched 22 new products in one year, 100% on time and on-budget, generating $50MM+ in revenue. The food storage launch — the S'well Insulated Food Container — became the fastest-selling product in the company's 10-year history at 100k+ units in year one.

It won the "Best of International Home & Housewares" Award in 2019. I also led a company-wide transition to sustainable packaging that drove $800k+ in annual cost savings and strengthened brand equity.

"Being the first PM isn't about having all the answers — it's about building the system that finds them. The consumer testing platform we built in month two is still how they validate products today."

What I owned

Market & consumer research — 200+ consumer insights, market sizing, and competitive analysis

Consumer testing platform — built from scratch; first in company history

Product development process — new XF framework, 30% faster timelines, CEO recognition

End-to-end GTM — 22 launches, 13 XF partners, SVP alignment, 100% on-time delivery

0 → 1 Consumer Goods New Category GTM First PM Consumer Research

Additional Experience

More context,
briefly.

Adobe · 2020

GTM Strategy — Content & Commerce

Conceptualized, proposed, and executed go-to-market for Adobe's combined Magento and Adobe Experience Manager offering. Built the financial model quantifying the market opportunity and trained 100+ salespeople across EMEA and North America.

$1.5Bn

Market opportunity modeled

100+

Salespeople trained

Procter & Gamble · 2014–2018

Global Pantene Innovation — $2Bn+ Business

Owned the 2019–2020 product roadmap for Pantene's global business across China, Europe, and the US. Led 152 launches across 23 global initiatives. Mitigated a supply chain crisis from new raw material regulations with zero launch delays.

152

Global launches

$2Bn+

Business managed

10k+

Consumer data points

Facing a similar challenge?

Let's talk about
your product.

30 minutes. No pitch. I'll tell you what I see and whether I can help.

Book a Free Strategy Call About Amy