eProcurement Toolkit

You are here: sourcing, stop 1/7

The lineStrategic sourcing

How do you reduce tail spend, step by step?

Short answer

You reduce tail spend by finding groups of similar small purchases in your data, giving each group one way to buy, and stopping new one-time suppliers from refilling the tail.

Cite this answereProcurement Toolkit, “How do you reduce tail spend, step by step?”, updated 14 Nov 2026, eprocurementtoolkit.org/strategic-sourcing/reduce-tail-spend

Key takeaways

  • Decide first what you want to reduce: the number of suppliers, the number of invoices and orders, the price paid, or the risk from unchecked suppliers. Each needs a different lever.
  • Most of the gain sits in a few groups of similar purchases bought from many suppliers. Group invoice lines by what was bought, not by ledger account, to find them.
  • Give each group one lever: use an existing contract, award a new contract after a quick quote, add a catalogue, use a card, name preferred suppliers, or stop the need.
  • The tail grows back unless new one-time suppliers need a reason and dormant supplier records are switched off.
  • Show the result with the same data a year later: suppliers and invoices removed, and the process cost they carried at your own cost per invoice.

How do you reduce tail spend?

You reduce tail spend in three moves. Find the groups of similar small purchases that many suppliers share, give each group one simple way to buy, and stop new one-time suppliers from refilling the tail. Much of the saving often comes from fewer suppliers, orders and invoices rather than from lower prices.

Tail spend is the many small, often one-off purchases that procurement does not manage directly. The CIPS glossary describes it as the bottom 20% of spend, which usually accounts for 80% of suppliers. What the tail is made of, with examples and types, is explained in what is tail spend. How to define it, the buying channels and the measures of a full programme are in tail spend management. This guide is the practical plan: what to do first, with a worked example.

What exactly are you trying to reduce?

“Reduce tail spend” can mean four different targets. Choose one or two before you start, because each needs its own lever and its own measure.

Suppliers
Fewer supplier records to create, check and keep current. Measured as the number of active suppliers in the tail.
Transactions
Fewer orders, invoices and payments to process. Measured as tail invoices per year. Each invoice carries a processing cost, so this target is often the one that saves the most.
Price
Lower prices on items bought again and again at list price. Measured on the same items before and after.
Risk
Fewer suppliers created in a hurry without checks, and fewer dormant records that can still be paid. Measured as one-time and dormant suppliers.

Reducing spend itself, the money, is a fifth target, and the hardest. It happens when a need is stopped or standardised: three teams cancel duplicate software subscriptions, or the company stops ordering branded gifts for every event.

What are the steps to reduce tail spend?

The steps below take a first cycle from data to results in about a year. They assume a spend analysis exists or can be run on twelve months of invoices, card transactions and expenses.

  1. Extract twelve months of tail spend under your written definition: suppliers, invoice lines, card transactions, amounts, sites and requesters.
  2. Group the lines by what was bought, using descriptions and supplier activity, not only the ledger account: “safety gloves” and “nitrile gloves” are the same group.
  3. Rank the groups by number of suppliers and number of invoices, then by spend. Pick the five to ten groups with the most suppliers for the same need.
  4. Choose one lever per group: an existing contract, a new contract after a quick quote, a catalogue, a card, a preferred supplier list, or stopping the need.
  5. Run the quick sourcing for the groups that need a new contract, and switch requesters to the new route with a short message and a catalogue entry.
  6. Close the door: require a reason before a new one-time supplier is created, and switch off supplier records that have not been used for a set period.
  7. Measure again on the same definition after six and twelve months, and start the next cycle with the next groups.

Step two is where most projects stall. Tail purchases are often coded to broad accounts such as “other services” or “sundry supplies”, so a ledger report shows no pattern. Reading the descriptions, even on a sample, finds the groups that the accounts hide.

How do you find the groups worth acting on?

The groups worth acting on are those where many suppliers sell the same thing to the same organisation. The fictional example below shows a company whose tail, under its own definition, holds 1,000 suppliers, 12,000 invoices a year and €4.5 million of spend. Grouping the lines by what was bought finds six groups.

Worked example (fictional): groups of similar purchases found in one year of tail spend, and the lever chosen
Group of purchasesSuppliersInvoices per yearSpend (€ thousand)Lever chosen
Lab and safety consumables bought by each site381,450610New contract after a quick quote
Office supplies and IT accessories bought outside the catalogue521,900340Redirect to the existing catalogue contract
Local repairs and maintenance callouts641,100520Preferred suppliers per region, with agreed rates
Training courses and conference fees85420260Purchasing card within a limit
Software subscriptions bought by departments47610390Cancel duplicates; move the rest to the IT contract
Printing and promotional items29330180Standard items from two suppliers
Six groups3155,8102,300
Rest of the tail (one-off purchases)6856,1902,200Card or quick quote; no group action
Whole tail1,00012,0004,500

In this fictional tail, six groups hold under a third of the suppliers but about half the invoices and spend: start there and leave the genuine one-off purchases alone.

The totals check: 38 + 52 + 64 + 85 + 47 + 29 = 315 suppliers, and 315 + 685 = 1,000; 5,810 + 6,190 = 12,000 invoices; €2,300 thousand + €2,200 thousand = €4.5 million. The six groups hold 315 of 1,000 suppliers (31.5%), 5,810 of 12,000 invoices (about 48%) and €2.3 million of €4.5 million (about 51%).

Which lever fits each group?

Each lever reduces something different. Choose the one that matches the target you set and the reason the group exists, and accept that some levers cut suppliers but not invoices.

Levers to reduce tail spend: when each fits, what it reduces and what to watch
LeverWhen it fitsWhat it reducesWatch out for
Use an existing contractA contract already covers the item, but people buy elsewhereSuppliers, priceThe contracted route must be as quick as the old one, or purchases drift back
New contract after a quick quoteA recurring need with no contract, bought by several sitesSuppliers, price, invoices if the supplier sends one monthly invoiceAgree delivery to each site and a monthly consolidated invoice in the terms
Catalogue or punchoutStandard items bought often in small quantitiesSuppliers, invoices, priceSomeone must keep the catalogue items and prices current
Purchasing cardLow-value one-off buys from suppliers used onceSupplier records, invoicesCard limits, blocked merchant types and monthly reconciliation
Preferred supplier listUrgent or local services where speed mattersUnchecked suppliers, price through agreed ratesInvoices stay one per job; the gain is in checks and rates
Stop or standardise the needDuplicate subscriptions, custom versions of standard itemsSpend itselfNeeds a business owner who agrees to the change
Outsource the tailToo many groups for the team to handleSuppliers and invoices seen by your teamThe provider’s fee, its rules and who checks its suppliers

Cards and catalogues cut invoices; contracts and preferred lists cut suppliers and prices; only stopping a need cuts the spend itself.

Organisations with very large tails now automate part of this work. Harvard Business Review described in 2022 how Walmart automated negotiations with tail-end suppliers, because hiring enough buyers to negotiate with each of them would have cost more than it returned. Automation changes who runs the negotiation; the choice of lever per group stays with procurement.

How do you source tail spend without a full tender?

Tail spend sourcing means buying a group of small, similar needs as one, with a light process. Add up the yearly volume of the group across sites, write one short specification, and ask three to five suppliers to price it with delivery to each site and one monthly invoice. A request for quotation is usually enough; a full tender is rarely worth it for a group of a few hundred thousand euros.

  • Aggregated quote. One request for the yearly volume of a group, such as gloves for all sites, priced per item with delivery terms.
  • Spot quote. For a one-off buy above the card limit, two or three quotes and a written record of the choice.
  • Call-off from an existing framework. If a framework agreement already covers the need, order from it instead of starting a new process.

Public buyers have extra rules and tools. In the European Union, Article 5(3) of Directive 2014/24/EU forbids splitting a procurement to stay below the thresholds unless objective reasons justify it, so grouping the tail can be a legal duty as well as a saving. The same Directive lets buyers place orders under a framework agreement, with or without a mini-competition among its suppliers (Article 33), or use a dynamic purchasing system, a fully electronic process for commonly used purchases (Article 34). In the US federal government, micro-purchases, up to $15,000 under FAR 2.101, may be awarded without competitive quotes when the price is reasonable, and FAR 13.201(b) makes the government purchase card the preferred way to pay for them.

How do you stop the tail from growing back?

The tail grows back because creating a new supplier is easy and switching an old one off is nobody’s job. Two controls stop that: a gate on new suppliers and a regular clean-up of dormant ones.

  • A gate on new suppliers. Before a one-time supplier is created, the requester gives a reason and sees the preferred suppliers for that need first. The supplier onboarding checks, including bank details, apply to every new supplier, however small.
  • A regular clean-up. Once a year, switch off supplier records with no order or payment for a set period, such as two years, so that they cannot be paid without a new check.
  • One owner for supplier records. A named team creates, changes and switches off suppliers; requesters and buyers cannot do it themselves. The supplier master data management guide explains the role.

Dormant records are a real risk, not tidiness. A 2025 internal audit of the City of Madison’s vendor master file found that 9,927 of its 19,629 active vendors had no recorded transaction for over four years, and that vendors marked inactive could still be used for invoice processing and payments. The city then moved 8,308 records to a status that blocks use. Fraud follows the same gaps: in the Association for Financial Professionals’ 2025 survey, covering 2024, 45% of respondents cited vendor impostor fraud, a form of business email compromise where someone poses as a supplier to divert a payment.

How do you show that tail spend went down?

You show it by running the same extract, with the same definition, a year later. The table below sets the targets of the fictional plan for the six groups, so that the result can be checked line by line.

Worked example (fictional): suppliers and invoices for the six groups, before and after the plan
GroupSuppliers beforeSuppliers afterInvoices beforeInvoices after
Lab and safety consumables3821,45024
Office supplies and IT accessories5211,90012
Local repairs and maintenance6481,1001,100
Training and conference fees (card)85142012
Software subscriptions47161012
Printing and promotional items29233024
Six groups315155,8101,184

The plan removes 300 suppliers and 4,626 invoices a year; repairs keep one invoice per job, so their gain is checked suppliers and agreed rates, not fewer invoices.

The totals check: 315 − 15 = 300 suppliers, and 5,810 − 1,184 = 4,626 invoices. Monthly invoices from one or two suppliers give 12 or 24 a year; card purchases arrive on one monthly statement, and the card issuer is the only supplier record left for training. To turn the invoices removed into money, multiply them by your own cost to process one invoice and pay it. APQC’s accounts payable benchmarks (2018) show that this cost varies widely: top performers spend about one-fifth of what bottom-quartile organisations spend per invoice, so use your own figure, not someone else’s.

Price savings are measured separately, on the same items before and after, and spend avoided by stopping a need is counted once, in the year it stops. Keep the three apart in the report, so that finance can check each one.

Questions

How long does it take to reduce tail spend?

A first cycle on five to ten groups usually takes several months to a year: a few weeks for the data and grouping, then the quick sourcing, the catalogue changes and the new supplier rules. Invoice and supplier counts show the change only after the new routes have run for a few months.

Can you reduce tail spend without new software?

Yes. A spreadsheet of twelve months of invoice lines is enough to find the groups, and most levers, such as contracts, cards, preferred suppliers and a rule on new suppliers, work in an existing ERP. Software helps to keep the grouping current and to route requests, which matters more in the second and later cycles.

Is cutting the number of suppliers always good?

No. Relying on one supplier for a need that stops production or service is a risk, and some public bodies have policies to buy from small or local businesses. Keep two suppliers where continuity matters, and check your supplier diversity commitments before consolidating a group.

What is a realistic target for tail spend reduction?

There is no reliable public benchmark, because organisations define the tail differently. Set the target from your own groups: for each one, the number of suppliers and invoices the chosen lever should leave, as in the worked example. The sum of those targets is a plan you can check.

New guides on the strategic sourcing line

One short email when a guide on this line is added or revised. Nothing else.

Coming soon: sign-up opens with the first email.

Seb Troz

Source-to-pay consultant · written and maintained this guide

Seb Troz is a source-to-pay consultant who implements Ivalua for buying organisations. He writes the guides on eProcurement Toolkit from that hands-on work.

Disclosure: the author works as an Ivalua consultant. Pages that compare software say so next to the comparison.