A model contract at Geldmaat & Brink's (2022)

Geldmaat manages a large number of coin-operated vending machines in the Netherlands, for which Brink's Solutions Netherlands provides cash transport and maintenance. Since the banks transferred their ATMs to Geldmaat and Brink's acquired the cash-in-transit companies in the Netherlands, both parties have been interdependent.

In 2019 and 2020, Geldmaat and Brink's negotiated a new contract for the next seven years. This proved challenging in an uncertain market with declining volumes annually (but no one knew exactly what the future would bring). Furthermore, a new coin-operated machine infrastructure was being rolled out, further increasing the uncertainty. When the parties failed to reach an agreement during 2020, Provoque Consulting was brought in for advice. SD&Co was subsequently engaged to develop the model.

The question from Geldmaat and Brink's to Provoque and SD&Co is twofold:

1. "Which activities and costs in the coin supply chain will be affected by a further decline in market volume, and what will the required capacity be?"

2. "What process allows us to annually determine the contract price for the coming year and manage the performance of the coin supply chain?"

During this project, we learned a number of general lessons that may be useful to other companies that (soon) have to renew their contract with a key supplier/customer and are facing a lot of uncertainty.

---

Coin Supply Chain

The coin supply chain consists of all the links between receiving coins from the Coin Acceptance Machines (MIAs; deposits by customers) and dispensing them again from the Coin Dispensing Machines (MIAs; customers withdrawing coin rolls). This includes retrieving the coins from the full MIAs (transport), counting and sorting the coins at the coin counting center, rolling the coins, and replenishing the coin rolls in the MUAs.

---

1. A model contract offers peace of mind in changing times

The new contract between Geldmaat and Brink's stipulates that a fixed price will be used in the first year, and that once the simulation model is complete, this model will guide the price for each subsequent year. This involves not (solely) following the model's output, but rather the price resulting from an intelligent discussion of scenarios and realistic objectives.

During the project, we saw a sense of calm emerge between the parties from the moment the contract was signed. No one knew exactly what the future held: the coronavirus pandemic, coin use had collapsed, the new coin-operated machines were being rolled out (no longer at banks, but at new locations), the new machines were still experiencing teething problems, and so on. For the first year, the price was clear, and there was confidence that the changes could be incorporated into the model. This provided peace of mind and confidence that a fair price would be agreed upon for future years, regardless of what that future would look like.

Implementing a Demand and Operations Planning (D&OP) process—a derivative of the Sales and Operations Planning (S&OP) process—has also contributed to this. By jointly reviewing volumes and activities at the tactical and strategic levels every three months, as well as the extent to which current capacity is still sufficient, they remain well-informed about changes and can discuss any desired adjustments in a timely manner.

Conclusion:

If the future is uncertain (in terms of demand, infrastructure, etc.), it's likely necessary for both customer and supplier to work together. A model contract with an accompanying consultation structure offers a solution for reaching the most fair possible long-term agreement.

2. Transparency, Trust and Traction are important conditions for a good customer relationship

At the start of the project, there was little trust between the two parties, and traditional customer-supplier relationships were the norm in the chain (between the banks and Geldmaat, between Geldmaat and Brink's, and between Brink's and Merlin), making it difficult to properly manage the coin supply chain. Only when we truly delved into the process with a team from Geldmaat and Brink's during the model's development did it become clear to those involved which activities and functions were involved in this coin supply chain (14 different functions at Brink's, to be precise). By making all activities and costs transparent and working together on this, trust gradually grew. When it ultimately turned out that the original estimates in Brink's' quote were quite consistent with the model's detailed analyses, this trust increased even further.

This increased confidence ultimately led to increased demand for process improvements . After the full rollout of the new coin supply chain—which is primarily located at GAMMA stores and no longer at the banks—the new coin machines malfunctioned twice as often, requiring a technician visit (previously, some malfunctions could be resolved by the bank clerk). As a result, incidents were reported 10 times as often, and calls were made 40 times (!) as often. This was discussed during the D&OP process in the contract agreements for next year. Both parties are now in discussions to explore how they can jointly reduce these numbers and define realistic targets for the coming year.

Conclusion:

Mutual trust isn't something to be taken for granted. By working transparently with each other and gaining insight into each other's processes, trust and the willingness to go the extra mile for each other are built.

3. Take time to analyze complex models

One of the takeaways we learned from this project is that complex models have a specific use case. Small models, for example, are ideal for collaboratively testing various scenarios ("what if...") with a group in the form of a workshop. Large, complex models require more time and focus. It's wise to divide the model work into several workshops: (1) a preliminary workshop in which you and the stakeholders jointly determine what you're going to test, and (2) a subsequent workshop in which you discuss the results with the stakeholders. With large, complex models, it takes more time to run scenarios and, especially, to analyze where differences come from. Experience shows that the quality of this analysis is better if you don't have eight sets of eyes watching your back (especially if some of those eyes are starting to get a little impatient).

Conclusion:

Take your time with the (interim) analyses of the model to ensure quality. Large, complex models take longer to analyze properly, so this is best done between workshops.

This article was written in collaboration with Willem van Oppen of Provoque Consulting.