E-Commerce

Case Study Enterprise Customer Support

Two ecosystems, one customer.

Whole Foods Market's Catering-on-Amazon platform lacked a cohesive, efficient support experience, making it hard to place and modify customer orders. As Amazon integrated ecosystems with Whole Foods, its customer service tools needed to meet Amazon's streamlined digital standard.

The unified agent console running on a desktop display

Our goal was to create parity between the two tool ecosystems, so customers and Customer Care finally meet in one place.

Building on what already works.

Amazon had already solved several of these problems elsewhere. The fastest route to parity was to find that work first, then design only what was genuinely missing.

01

Components that had already earned their place.

We identified proven Amazon components that already solved common customer issues and could adapt to Whole Foods catering workflows, so development effort went into the gaps rather than into rebuilding what existed.

Existing Amazon interaction patterns adapted for Whole Foods catering workflows
Established Amazon components, reused for common support tasks.
02

Then the part nobody had drawn.

Working with developers, project managers, and end users, we mapped the current-state tool ecosystem end to end. Seeing every channel and every tool on one wall is what turned a vague sense of friction into a specific list of gaps.

Less ownership and control More ownership and control
Global support activities
In-store back of house
In-store front of house
Journey
Assortment & recipe
Data
Production planning
Execution
Staging & stocking
Handoff & reconciliation
Problem
What do we want to sell?
Is the item set up everywhere?
How do we make it?
What does it cost us?
Where do we buy it from?
What does the customer pay?
When can they buy it?
How much will we need?
Do we have it in the store?
How much can I sell today?
What do I need to make?
Is it safe to sell?
How much does it weigh?
Where do I put this?
How do I close the order?
What needs refilling?
Is the count still right?
Step
Planning
Item setup
Recipe setup
Cost setup
Sourcing
Price setup
Availability
Forecasting
Ordering
Allocation
Prep
Temp check
Labeling
Staging
Payment
Replenishment
Counting
Process
Assortment plan
Catalog entry
Recipe build
Cost load
Purchase order
Price publish
Window publish
Demand plan
PO creation
Inventory allocation
Production run
Safety check
Label print
Order staging
Reconciliation
Restock
Cycle count
Current tooling by channel 102 touchpoints across 6 channels
Setup happens three different ways
No shared view of what is actually available
This part already works
Owned storefront
Marketplace
Third-party delivery
Phone & walk-in
Corporate catering
Subscription & pre-order

Duplicated 15 Gap or unknown 18 In place 53 Works well 6 Not applicable 10

Current tooling by channel, mapped against the journey it has to serve. Names and systems are scrubbed; the structure, and the state of each touchpoint, are as we found them.

Scroll the map sideways to read it.

The map is a table of six sales channels against seventeen journey steps. Each cell names the kind of tool used and its state: in place, works well, duplicated, a gap or unknown, or not applicable. Totals for each state are given in the key below the map.

The solution

One platform, built to Amazon's standard

A unified, enterprise support platform that brings Whole Foods catering onto Amazon's digital standard β€” giving agents every action, every order, and the full customer context in one place.

Customer issue context, order-match cards, and consistent Amazon interaction patterns replace the fragmented tool-switching that slowed teams down.

The outcome

Parity, not patchwork

The unified platform gives Amazon Fresh and Whole Foods Market Customer Care a single, consistent place to work, aligning two ecosystems around one seamless support experience.

One place for every action

Agents locate orders, see full customer context, and resolve issues without switching between disconnected tools.

Built on proven patterns

Reusing established Amazon components kept the experience familiar for agents and reduced duplicated development effort.

Grounded in real research

Every design decision traced back to stakeholder workshops, current-state mapping, and the pain points teams named directly.

Next: me@slalom