Most Procurement Teams Have Never Sourced Their Bank

TL;DR: Every supplier gets a tender. The foreign exchange spread on cross-border payments gets a rate card nobody read. Group companies move funds to subsidiaries at the bank's standard rate, month after month, and the cost never appears as a line item because it is buried in the rate. Procurement can source this like any other category, but only in partnership with treasury: thresholds, volumes, timelines and existing conditions first, benchmark second. Procurement Union is building FX & Payments as a new Companion category to do exactly that.

The spend nobody owns

Ask a finance director who negotiated the FX terms on the company's intercompany transfers and the honest answer is usually: nobody. The bank relationship was set up years ago for lending, cash management or payroll. FX came bundled. The rate applied to each transfer is the bank's standard spread over mid-market, and the invoice never shows it, because there is no invoice. The cost is the difference between the rate received and the rate that was available, and that difference is invisible unless someone goes looking.

We have seen this in companies that run tight procurement on everything else. The parent transfers operating capital to a subsidiary every month. Regional entities pay head office for shared services. A holding company funds a new market. Each transfer is small enough not to trigger a review and large enough, in aggregate, to matter.

A worked example: 150 basis points of spread on USD 20M of annual cross-border flows is USD 300,000 a year. That is a supplier contract most procurement teams would run a full sourcing event for. Here it was never put in competition because nobody saw it as a supplier contract.

Why procurement does not touch it

3 reasons, and they are all fixable.

• It looks like finance territory. FX sits with treasury. Procurement teams assume the bank is a relationship, not a supplier, and stay out.

• There is no spend data. Spend analysis runs on invoices and purchase orders. FX cost sits inside the exchange rate, so it never shows up in the cube. If it is not in the data, it is not in the category plan.

• The incumbent is comfortable. Changing a banking relationship is painful, and nobody is proposing that. But that is not the only option, and the alternatives rarely get evaluated because the question is never asked.

Treasury is the stakeholder, not the obstacle

This is the part that decides whether the exercise succeeds. The wrong way to do it is to benchmark the bank's rates in secret, write a memo for the CEO, and present treasury with a finding. That ends the conversation.

The right way is to treat treasury as the category stakeholder from day one, the same way procurement would treat IT on a software renewal or facilities on a cleaning contract. Nothing is confidential. Nothing is "for CEO eyes only". The sequence is:

1. Get the thresholds, volumes and timelines in advance. Which corridors carry volume, how much, how often, and when the next transfers are due. Treasury has this; procurement does not.

2. Understand the conditions already in place. Advance booking arrangements, payment terms, special corridors with negotiated rates, hedging positions, forward contracts, any relationship pricing tied to lending. A benchmark that ignores these will be wrong and will cost credibility.

3. Agree what good looks like. Spread in basis points over mid-market at the time of each transfer, all fees included. That is the price. Everything else is packaging.

4. Then benchmark. With the data treasury provided and the conditions treasury explained.

   
   

Treasury owns the relationship. Procurement knows how to put a price in competition. Neither can do this alone.

What the benchmark actually measures

The calculation is simple and the discipline is in the inputs.

• Rate received on recent transfers, from the bank statement, not the rate card.

• Mid-market rate at the timestamp of each transfer.

• Spread is the gap, expressed in basis points. Add any explicit fees.

• Annualised cost per corridor, using treasury's volume figures.

  
  

These are illustrative. The point is not the band; it is the gap between what the company pays and what the same volume would cost with a negotiated rate from the incumbent, a second provider, or a different structure.

Three options, in this order

1. Renegotiate the incumbent. Walk into the bank with the benchmark and the volume. Most relationship banks will move on spread for a client that asks with data. This is the first option because it costs nothing to execute and preserves the relationship treasury has built.

2. Add a second provider. For specific corridors where the incumbent will not move, route that flow elsewhere. Dual-sourcing is standard procurement practice. It also gives treasury a live reference rate for the next renegotiation.

3. Restructure the intercompany flow. Often the biggest saving, and it needs no new supplier. Net payables and receivables between entities before converting anything. Batch transfers monthly instead of ad hoc. Review the currency of intercompany invoicing so fewer conversions happen at all. This is a treasury decision with a procurement-style analysis behind it.

The new Companion category: FX & Payments

Procurement Union is building FX & Payments as a new insight category inside the Companion. It treats the bank as a supplier and the spread as a price.

The user enters corridors, volumes, provider and the rates actually received. The Companion benchmarks each transfer against mid-market at the time it happened, annualises the cost per corridor, and produces a one-page brief that procurement can take into treasury: what the company pays today, what comparable volume costs across provider types, and the three options above with the incumbent renegotiation listed first.

Where the community has submitted anonymised spread data for a corridor, the benchmark uses it. Where it has not, the bands are labelled as illustrative. The Companion advises on the sourcing; it does not execute transfers and does not recommend a provider over another without showing the alternatives.

Few procurement people get involved in this category. The ones who do find money that was already being spent.

Procurement Union builds the self-hosted AI Companion for procurement professionals. FX & Payments joins the existing insight categories in the next release.