Runderkamp: from fragmented pricing data to one reliable basis for margin and growth

About Runderkamp

Runderkamp is a food retail and catering business with multiple butcher locations, a central production environment and a catering operation. Within this mix, pricing is business-critical. Changes in purchase prices have to be translated quickly and correctly into sales prices, while Runderkamp wants to keep a grip on margins, commercial exceptions and the balance between production and shops.

That grip had become increasingly difficult because the pricing process had drifted, over the years, across different systems, files and personal ways of working.

The business challenge

The Calculatie winkelprijzen workbook had long been the heart of the pricing process. Information from supplier lists, invoices, webshop checks, EasyParty, QlikView, Digi @Fresh and various Excel files was brought together in it by hand.

Before a price change could be applied, data had to be collected, linked, checked and sometimes reinterpreted. Suppliers and articles appeared under different numbers or names. The same PLU could mean different things in different contexts. Price and margin rules lived partly in formulas and partly in the heads of employees.

As a result, the process was labour-intensive, person-dependent and hard to trace. A technically correct dashboard placed on top of that situation would not have solved the underlying problem.

First the basis, then automation

Impacture deliberately did not start with a dashboard, chatbot or AI model.

Automation amplifies what is already there. When source data contains different definitions, numbers and quality levels, automation spreads the errors and inconsistencies faster too. Reliable BI and AI are therefore only possible after data has been:

centralised → standardised → optimised → automated

Impacture built one managed data environment for Runderkamp in Microsoft Fabric.

Through connectors and controlled import processes, data is pulled from, among others:

• EasyParty;
• Digi @Fresh and the existing QlikView information;
• Excel and XLSX files;
• supplier and article files;
• gross margin targets;
• additional external source information.

Where an API is available, it is used. Where systems or suppliers only deliver files, those are processed through controlled imports. Runderkamp therefore is not tied to a single technical integration model.

From raw data to reliable business information

The solution is built on the Medallion architecture.

Bronze — capture

All delivered source data is stored unchanged, together with source, file, batch and date metadata. It therefore remains visible what was received and when.

Silver — standardise and control

Products, PLUs, suppliers, categories and prices are cleaned up and linked together. Duplicate or conflicting data is resolved. Price and margin rules are applied consistently. Data that cannot be reliably linked does not disappear — it ends up in a visible exceptions and alerts flow.

Gold — steer

Only validated, business-ready data is included in the semantic model for reporting, analysis, operational decision-making and AI and BI.

The full environment is set up according to DTAP: Development, Test, Acceptance and Production. New connections, calculation rules and reports are therefore developed and tested in a controlled way before they become available in production.

It used to take a lot of time to figure out which file, number or price was authoritative. Now deviations become visible automatically and we only assess the exceptions that really need our attention. — Cor Runderkamp (proposed quote)

Sharing securely via the Impacture Sharing Engine

The information is made available through the Impacture Sharing Engine. Users sign in through a secure login and gain access only to the data intended for their role or audience.

Row-level security then determines, at data level, what someone is allowed to see. Management can have the overall view, while employees only get access to the products, locations or processes relevant to their work. The same mechanism makes it possible to share selected insights with external partners or suppliers without exposing internal information or the underlying environment.

Reports are delivered as securely embedded views. There is therefore no need to distribute separate files, spreadsheet copies or per-audience report versions.

The result

Runderkamp has one reliable basis for product, supplier, price and margin information.

Price updates are processed faster and more consistently. Supplier-specific price rules are applied in a controlled way. Historical prices remain available and every change is traceable. Incomplete or conflicting links are made visible, so they can be resolved deliberately rather than ending up unnoticed in reports.

Employees spend less time collecting and reconstructing information. Management and operations work from the same definitions and the same current data. At the same time there is room for controlled commercial exceptions: automation supports Runderkamp's craft, but does not replace it.

The central, standardised data environment also forms a reliable basis for further applications, such as automated alerting, price and margin analysis, price elasticity research and natural-language questions.

Runderkamp has therefore not just automated a pricing process. The company has created a scalable information basis on which future reporting, automation and AI can be built responsibly.