Guinet-Derriaz has been extracting and working stone since 1912. In 2022, the group has fourteen quarries, around ten million euros in revenue, customers abroad, and no information system. Orders live in spreadsheets kept by each team, and no question that crosses two teams has a single answer. I am asked for tools, one for the sales team, one for the quarries, quickly.
One point of margin on ten million euros is worth a hundred thousand euros: the budget of the engagement.
The same stone, two unrelated names
The problem underneath the tools: a thousand or so commercial names, chosen sale by sale, unrelated to the process that makes the pieces. The salesperson sells a name, the workshop produces a sequence of steps, and nobody knows which steps hide behind a sales name. Prices are set by hand, with no view of the margin, and production gets it wrong.
- A commercial name one among a thousand
- A price set by hand
- Stone a cost
- Selection a cost
- Thickness a cost
- Finish a cost
- Format a cost
no link between the two
The normalised reference links the two. Its name says how the piece is made, and each attribute carries its cost.
Cost price, the sum of the steps. Selling price, the costs plus the margin. For the first time, one follows from the other.
In this order, and not the reverse
- Normalise the catalogue, before choosing a single tool. Months on reference data nobody had asked for.
- Test the processes in Excel and Airtable, with the teams. A spreadsheet remains the right answer as long as a process has no stable rule.
- Choose the vendors for sales and operations, only then. Buy rather than build, because reference data done first makes every following tool small.
The tools the teams were waiting for came last. A tool built on four versions of the catalogue would have failed at the first disagreement between two teams, and sent everyone back to their spreadsheets. I had to hold this order in front of people who wanted to see something.
What came of it
By the end of the year, the group was quoting, tracking its orders and delivering from shared reference data instead of a dozen private files. Sales, operations and the quarries were working on the same data. I have no measure of margin or production errors after the engagement. The reference data and the processes lived in Excel and Airtable: there is no screen to show, and the five attributes above are the ones I remember.
The group is the second customer from the stone industry, after Chunk, which had put me in touch.