Case

Laudrup Vin – Data analyses that changed course

Laudrup Vin is a Danish wine importer founded in 1993 and part-owned by former footballer Michael Laudrup together with Per Buhl, who is managing director. The company imports wines from around the world and complements the range with…

Anders Qwist23 July 2026, 2 minute read

Laudrup Vin is a Danish wine importer founded in 1993 and part-owned by former footballer Michael Laudrup together with Per Buhl, who is managing director. The company imports wines from around the world and complements the range with selected gastronomic products.

“Working with Anders has made our decisions far more data-driven. Last year’s figures are still a reference, but today we steer primarily by budget and the underlying customer economics. That has given us greater confidence to invest offensively – with markedly better results on both top line and earnings.”

– Carsten Vinter, CFO, Laudrup Vin

Challenge

Laudrup Vin has a solid market position and a strong range. Even so, there was a sense that the e-commerce channel could contribute more to the bottom line and become a more significant part of the overall business. The question was therefore fundamentally strategic: was e-commerce a channel that should be scaled down – or was the challenge rather that the channel “simply” needed optimising to contribute better to the future business?

The initial hypothesis was that the company was spending disproportionate resources on the webshop relative to the result, and that the marketing investment should be better targeted to raise the return.

Solution

To create a solid basis for the decision, the e-commerce channel was analysed with a focus on channel contribution, customer economics and scalability.

The analysis had four central elements:

  • Repeat purchase rate (annual)
    To assess Laudrup Vin’s ability to build loyalty, and with it a base of revenue that does not depend directly on paid advertising.
  • Customer Acquisition Cost (CAC)
    To map what it actually cost to acquire a customer who had not bought from Laudrup Vin before – and to set that against revenue and earnings per order.
  • Customer economics (value over time and allocation of fixed costs)
    To identify the real earnings per customer over time (given the insight into repeat purchase rate) plus the allocation of fixed costs such as salaries and software, in order to assess the overall channel economics — including under future growth, where the mix of fixed and variable costs changes.
  • Permission efficiency
    To analyse the difference in revenue and profitability between customers with active permission, for example a newsletter subscription, and customers without – and to assess which segments actually contributed positively.


The analyses showed a markedly different picture from the original assumption. Laudrup Vin had:

  • A high repeat purchase rate, indicating strong loyalty and a good customer experience
  • A low CAC relative to both revenue and earnings per order
  • A very positive customer economics, once fixed costs were allocated correctly


The central conclusion was therefore clear:
The problem was not that too much was being invested in marketing – but that too little was.

On that basis, simulations were built showing the effect of increasing the marketing investment significantly. The simulation accounted for both new and existing customers, and for a conservative expectation of rising CAC.

Results

The simulations showed considerable potential for growth and earnings — in revenue growth, but above all in earnings.

On the strength of this, Laudrup Vin chose to:

  • Increase the marketing investment with a focus on scalable customer acquisition
  • Build a performance tool that continuously assesses the pull of the marketing effort and the campaign response, and makes it possible to adjust investment up or down based on actual results
 
 
 

Do you want to know which products carry your business?

Write a few lines about what you sell, and I will come back with a suggestion for how the four analyses can be run on your own data.