Case Studies

Helping shoppers find the right product in the right place

Turning millions of product-location combinations into a simple in-store experience

Sector

Retail and e-commerce

Client

Leading Benelux home improvement retailer

Challenge

Help customers find individual products across hundreds of different store layouts

Solution

End-to-end product-location platform connecting enterprise product data, local store knowledge, mobile tooling and customer-facing experiences

/

/

Helping shoppers find the right product in the right place

/

/

Helping shoppers find the right product in the right place

01

The challenge

“Where can I find this product?”

It is one of the simplest questions in retail.

At scale, it becomes a serious data and experience problem.

The retailer operates more than 330 stores across the Benelux. Individual stores can carry tens of thousands of products, and every location has its own layout, aisles and shelving logic.

A product available in Store A might sit somewhere completely different in Store B.

Customers regularly needed help finding products.

Store employees spent significant time answering those questions.

And headquarters could not simply create one universal store map because local teams knew their own stores best.

The challenge was to make product location easy for customers without creating an enormous maintenance burden for employees.

02

What needed to work together

The customer experience looked simple:

Select product → see where it is in the store.

Behind that interaction sat several harder problems.

The system needed to know:

  • Which store the customer had selected

  • Which products were available there

  • How that store was organised

  • Where each product belonged

  • How locations should be described to customers

  • How changes could be updated quickly

  • How information should reach websites, apps and other channels

The project was therefore as much a data and operational challenge as a UX challenge.

03

What we built

One product-location platform for stores, employees and customers.

WeAreBrain designed and built an end-to-end platform connecting four layers.

Store knowledge

Local teams define how locations are described.

Product data

Enterprise product information is connected to individual store locations.

Employee tooling

Store staff can create and maintain mappings while walking through the store.

Customer experience

Location information becomes available across digital channels.

Together, these turn highly local store knowledge into reusable structured data.

1. Let stores describe themselves

The people working in the store know the layout best.

A centralised approach would have required headquarters to maintain hundreds of unique store layouts.

Instead, we gave local teams control.

Stores can describe locations in language that makes sense in that particular environment.

For example:

  • Aisle 17, left side

  • Back corner of the store

  • Drive-in, rear section

This creates location descriptions that are meaningful to customers rather than forcing every store into the same abstract structure.

2. Connect every product to a location

330 stores create millions of combinations.

Product data was already available centrally.

Store-location knowledge was local.

The platform connects those two worlds.

We built a backend integrated with the retailer's SAP environment that applies the business logic and transformation needed to connect products to individual store locations.

The result is a dynamic database containing more than:

04

17 million product-location combinations

across:

05

330+ stores

This becomes the foundation for every customer-facing product-location experience.

3. Make maintenance easy for employees

A good customer experience depends on good internal tooling.

The data would quickly become unreliable if updating it required complicated administration.

Store employees already carried Zebra handheld devices as part of their daily work.

We built a dedicated Android application for those devices.

An employee can:

  • Walk through the store

  • Scan a product

  • Select its location

  • Save the mapping

  • Continue to the next product

The system is designed so every individual item does not need to be scanned manually.

A store can complete the initial location setup in approximately 3 to 4 hours.

4. Put the answer where the customer needs it

Once the underlying data exists, it can be used across several customer journeys.

Product pages

When customers select their preferred store, the product page can immediately show where that item is located.

Location updates become available across the website and mobile applications.

Mobile apps

Customers can check product locations while planning or completing a store visit.

The same platform powers both web and app experiences.

Shopping lists

Customers can create lists of products they need.

The system can use those products and store-location data to help organise the in-store journey around the relevant aisles.

Conversational product discovery

A conversational interface was later added on top of the same data foundation.

A customer can ask:

“Where can I find wall plugs?”

The system can identify the relevant product and return the exact location in the selected store.

The important part is what sits behind the conversation.

The AI does not need to guess.

It has structured product and location data to work with.

06

AI works better when the data underneath is good

The chatbot was the final layer, not the starting point.

It would have been relatively easy to build a conversational interface that answered general questions about products.

It would have been much harder for that interface to reliably know where an individual product was located in hundreds of different stores.

The project therefore started with the underlying information architecture.

Store locations became structured data.

Products became connected to those locations.

Employee workflows kept that data current.

Only then could AI provide a useful conversational layer.

That pattern applies far beyond retail.

Good AI experiences depend on good systems underneath them.

07

Supporting store employees

The platform was designed for customers, but it also helps employees.

New members of staff can find products more easily while they learn the store.

Employees preparing click-and-collect or reservation orders can locate items faster.

Teams spend less time answering simple location questions.

The same data therefore supports both customer experience and store operations.

08

The result

One data foundation. Several useful experiences.

The platform transformed store-location information from local knowledge into a reusable digital capability.

17 million+ mappings

Individual products are connected to their physical location across hundreds of stores.

330+ stores

One platform supports hundreds of different local layouts.

3–4 hour store setup

Local teams can create the mapping without a lengthy central implementation project.

Multiple customer channels

The same location information powers websites, apps, shopping journeys and conversational search.

Useful internal tooling

Employees can maintain the underlying data as part of normal store operations.

09

Why this matters

Great digital experiences often depend on invisible operational design.

To a customer, the feature is small:

“Your product is in aisle 17.”

Creating that answer reliably required:

  • Enterprise data integration

  • Local store tooling

  • Millions of data relationships

  • Mobile development

  • Business logic

  • Web integration

  • App integration

  • AI

The complexity sits behind the experience.

The customer gets the simple part.

That is good product design.

AI services used

Product Design

Designing the complete journey across customers, store employees and multiple digital channels.

Software Engineering

Building the backend platform, mobile tooling and integrations required to make the experience work at scale.

Data Infrastructure

Structuring millions of product and store-location relationships so they can be reused across applications and AI.

Avatar-1
Avatar-2
Avatar-4
Avatar-4
Avatar-3

Have complex data hiding behind a simple customer problem?

That is often where the most useful digital products begin. We help connect data, internal tooling and customer experience into one system.