Tail Spend Policy: Setting Thresholds and Approvals

A tail spend policy defines which order goes through which path before a department decides on its own. It assigns every request to one of three paths: catalog, framework agreement, or a single creditor for everything else.

Add to that de minimis thresholds, staggered approval levels, and a clear list of permitted ordering paths. The policy only becomes effective once it is faster than the workaround around it. Otherwise departments keep ordering around procurement and the approval process.

At a Glance

• A tail spend policy assigns every order to one of three paths up front – catalog, framework agreement, or a single creditor for everything else.
• De minimis thresholds and approval levels only work when they scale with order value, instead of requiring the same path for every order.
• A policy only holds up if it is faster than the way around it – otherwise you get maverick buying.

Which Order Goes Where: Catalog, Framework Agreement, or Creditor

The first job of a tail spend policy is assignment, not control. Every request must be mapped to one of three paths in advance, without a department deciding case by case. Otherwise chance decides.

Catalog covers plannable standard needs: office supplies, IT accessories, recurring consumables. Framework agreement works where a fixed supplier delivers regularly, for instance maintenance parts for a specific piece of equipment. Both paths assume a known supplier who is already set up in the system. New cases do not fit either one.

For everything else, only the third path remains: a single creditor for needs with no fixed provider. Spare parts, custom-made items, trade-show supplies – cases that fit neither a catalog nor a framework agreement, because the right supplier is different every time. How much frequency and irregularity it takes for this third path to pay off is laid out in detail by the decision tree on outsourcing versus internal optimization.

If the policy skips this assignment, each department decides for itself when in doubt – and, when in doubt, onboards a new supplier instead of using an existing creditor. The same effort, twice over.

Staggering De Minimis Thresholds and Approval Levels Correctly

A single approval level for every order value either burdens small amounts with unnecessary review or leaves large ones with too little control. A policy only becomes effective once it scales by order value.

A sensible structure looks like this: below a defined de minimis threshold, approval from the requester alone is enough, with no further instance. Above that, up to a mid-range threshold, a single approval from the team lead suffices. Only above that threshold does the full approval chain kick in, involving procurement leadership and, depending on the amount, budget ownership from another department. Three levels are usually enough.

This staggering follows the same principle as other approval processes in the company: review effort should match actual risk, not the number of orders. For an overview of common approval structures, see the Haufe Akademie process glossary.

Without this staggering, every order, whether €80 or €8,000, ends up in the same review loop. That time is then missing elsewhere. Procurement teams already spend nearly three-quarters of their time on operational processing rather than strategic procurement, according to a CPO survey (Deloitte Global CPO Survey 2021).

Which Ordering Paths the Policy Allows

A policy without defined ordering paths stays theoretical. Paper alone helps no one. It has to specify exactly how an order gets placed on each of the three paths, otherwise every department finds its own informal channel.

For catalog and framework agreement, that's usually the existing order process in the ERP system. For the single creditor, a deliberately short list of permitted paths pays off: upload a quote PDF, paste a product link, or select from a private, contractually bound catalog. Three paths, no exception list on the side. More paths mean more exceptions, and every exception undermines the clarity a policy is supposed to create.

It also matters who is allowed to use these paths day to day. If the policy doesn't define that, every department ends up deciding for itself, again, which channel it considers fast enough. That is exactly what the policy is meant to prevent.

Why the Policy Has to Be Faster Than the Workaround

A policy that is correct on paper but takes longer in practice than the informal workaround does not get followed. This is the real test: if the regulated path is faster than maverick buying, the policy holds itself. If it's slower, the department bypasses it as soon as the need is urgent enough.

This applies especially to the third path, the single creditor. If approval there takes longer than a spontaneous order to a known contact outside the process, the policy loses its effect, even if it's correctly designed on paper. Speed beats any rule, however cleanly it's written.

A production department needs a replacement gasket for around €90 on short notice, an illustrative example. If the policy's de minimis threshold is unclear, the order waits on an approval that nobody actually wanted to review anymore. It would be faster to call the known contact from the last repair directly – exactly the behavior a policy is supposed to prevent. Not out of bad intent, but out of time pressure.

That's why a response time belongs in every tail spend policy, not just an approval chain. An order confirmation within 24 hours gives departments a reason to use the official path instead of going around it. Without that time frame, even the clearest assignment of catalog, framework agreement, and creditor remains a rule that loses out day to day.

The Tail Spend Policy on One Page

Nobody reads a policy in full if it needs more than one page. Eight points are enough for a document that actually gets used:

Scope. Which categories and order types the policy covers, and which it explicitly does not (direct material stays out of scope).

Classification. The rule that assigns an order to catalog, framework agreement, or the single creditor.

De minimis thresholds. The concrete euro thresholds above which a higher approval level applies.

Approval levels. Who approves at which value tier, from the requester up to procurement leadership.

Permitted ordering paths. The definitive list per path, with no informal exceptions on the side.

Response time. The committed time to order confirmation, so the official path stays the fastest one.

Ownership. Who maintains the policy and decides on edge cases.

Review cycle. A fixed date on which thresholds and approval levels are checked against actual ordering practice.

You don't need to reinvent this document every year. An annual review is usually enough, unless order volumes on the third path shift significantly.

Who Keeps the Policy Running Day to Day

A policy defines where an order belongs. That's enough. What remains open is who actually serves the third path – the single creditor – fast enough to keep the response time from the policy. Someone has to do it day to day.

Pedlar operates this model as a managed service. The 24-hour order confirmation that your response-time rule relies on is then actually backed up, instead of just existing on paper.

For how the four fundamental tail spend management models fit together, see the guide on the topic.

Frequently Asked Questions

How Often Should a Tail Spend Policy Be Revised?

An annual review is usually enough. If order volumes on the third path shift significantly, for instance because of a new department or a site relocation, an earlier look at the thresholds is worthwhile.

Does Every Category Need Its Own De Minimis Threshold?

No, a single staggered threshold across all categories is usually enough. An exception is only worthwhile when one category carries significantly higher risk than the rest, for instance safety-relevant spare parts.

Who Should Set Up a Tail Spend Policy Within the Company?

Usually procurement leadership together with accounting: procurement from the approval side, accounting from invoice review. Departments should know the ordering paths before the policy takes effect.

Does a Policy Prevent Maverick Buying Completely?

Not by its existence alone. It only prevents maverick buying when the regulated path is actually faster than the way around it – response time matters more than the wording of the policy itself.

Find out how the Single Creditor Model backs your tail spend policy with a reliable response time: Book a call →