Spend management: the complete guide for finance and procurement teams
Every business now spends money in more places, with more suppliers, than its finance team can see. This guide covers how to govern that spending, how to pay for it properly, and how to prove it afterwards.
Definition: What is Spend Management?
Spend management is the practice of controlling all the money a business spends with third parties, from contracted supplier costs to everyday employee purchases. It covers setting policies, issuing payment methods, tracking transactions in real time, and reconciling them with accounts. It is broader than expense management, which handles only employee claims.
How spend management fits next to other forms of business money management
Most businesses do some parts of spend management well, but not others. Contracted supplier spend is usually sourced carefully and paid against an invoice. A £30 taxi, a new design tool licence or a set of replacement laptop chargers tends to go on somebody’s personal card and surface as a claim three weeks later, if it surfaces at all.
Spend management treats both as the same problem. The money leaves the business either way. Either way it needs to be authorised before it goes, visible while it happens, and evidenced afterwards.
| Term | What it covers | Where it sits |
|---|---|---|
| Spend management | All third-party spending, end to end: policy, payment, evidence and reconciliation | The parent. Everything else sits inside it. |
| Expense management | Costs incurred by employees, and the claims process that reimburses them | A subset, focused on employee-incurred spend |
| Procurement | Sourcing, supplier selection, contracts and negotiation | A subset, focused on how you buy |
| Accounts payable | Receiving, processing and settling supplier invoices | A subset, focused on settlement |
| Travel and expenses (T&E) | Travel, subsistence, mileage and client entertainment | A spend category, usually managed inside expense management |
Why spend management matters now
Company spending is growing across every category we measure, and it is growing faster than the finance teams asked to keep track of it.
Soldo’s Procurement Spend Index compared the first half of 2025 with the same period in 2024, using anonymised spending data from over 25,000 customers across the UK, Italy, France, Spain, Germany and the Netherlands. Operational spending rose 26%, with everyday shopping up 33%. Software spend grew 43% in the UK and 37% in Italy. Professional services rose 59% in the UK and 41% in Italy. Travel and expenses rose 12%. Investment in AI tools rose 130% among large businesses.
Two things follow. The number of individual transactions is rising, and so is the number of suppliers behind them. Neither is matched by growth in finance headcount, which means the gap between what a business spends and what its finance team can actually see is widening rather than closing.
There is a cash position underneath this. 62% of UK small businesses surveyed are owed money from unpaid invoices, so the working capital available to absorb uncontrolled spending is thinner than the growth figures on their own suggest.
26%
Operational spending
43%
UK software spend
59%
UK professional services
The gap between business spending and what the finance team can actually see is widening
The Association of Certified Fraud Examiners' 2024 Report to the Nations estimates an average loss of £145,000 per expense fraud incident, with 20% of cases exceeding £1 million. The same research found expense reimbursement schemes run for an average of 18 months before being detected. In a separate 2024 survey, nearly one in four employees admitted to passing personal purchases off as business expenses. Manual processes do not just miss fraud, they create the conditions for it.
The five ways spend management could fail
No spend visibility
The dashboard shows what happened, not what is happening. Card statements arrive monthly, invoices arrive on supplier terms, and claims arrive whenever the employee gets round to it. By the time finance can see a month’s spending, the month is over and the money has gone. Decisions get made on data that is structurally out of date.
Outdated petty cash processes
Petty cash, a shared corporate credit card, blanket purchase orders and procurement cards were all meant to make buying easier. Each created a new problem. P-cards apply their controls only after the money has gone. Blanket purchase orders blur what was actually bought. Cash leaves paper receipts and no traceable record at all.
Manual, error-prone work
Managers waste 30 hours a month chasing receipts and approvals. Employees spend 18 hours a month on tail spend. Processing a single purchase order can cost between £50 and £1,000 once review time is counted, which is frequently more than the thing being bought. The work is not just slow, it is expensive relative to the value it controls.
Strain on employees
When the default way to buy something for work is to pay for it yourself and claim it back, the business has moved a cash flow problem onto its staff. 55% of UK employees hesitate before submitting a claim. Around one in five claims is rejected. Nearly one in four employees say expense claims are stressful, and in the UK challenging a rejected claim ranks just behind asking for a pay rise.
Fraud and audit exposure
Retrospective checking is a weak control. Card sprawl makes spending hard to attribute, duplicate subscriptions accumulate unnoticed, and off-contract buying leaves no supplier record. 74% of UK finance leaders now worry about AI-generated fake receipts, against 41% in Italy. A process that verifies documents after the fact is the process most exposed to a convincing forgery.
The confidence gap
Soldo surveyed 1,010 UK employees and 650 Italian employees alongside 250 finance leaders in each market between August and September 2025. The results describe two organisations that believe they are one.
69% of UK finance leaders believe their employees understand the expense rules. Only 39% of UK employees say they know exactly what can be expensed. 55% hesitate before submitting a claim. 41% buy something for work and never claim the money back. 38% avoid spending altogether, even where the spending would have helped the business.
That last figure is the expensive one. It is not a compliance problem and it is not an admin problem. It is a business quietly deciding not to do things it wanted to do, because the process for paying for them was unclear.
When employees are unsure, they do not ask finance. 46% ask their line manager and only 14% ask finance directly, which means policy gets interpreted through managers rather than applied consistently. The rules on paper may be perfectly clear. What matters is the version being enforced in practice, one conversation at a time.
| Figure | What it measures |
|---|---|
| 69% | Leaders who think the rules are clear |
| 39% | Employees who know what can be expensed |
| 41% | Buy for work and never claim |
| 38% | Avoid spending altogether |
The cost of doing nothing
Nothing about an unmanaged spend process is stable. Transaction volume rises, supplier count rises, the share of spending that never reaches the books rises, and the audit trail thins. The cost is not one number on one line. It is finance time spent reconciling instead of forecasting, procurement time spent processing instead of negotiating, and employees quietly absorbing costs the business never sees.
The anatomy of company spending
Two ways to classify spending, and why using only one of them is what makes tail spend so hard to control.
Most guides to spend management sort company spending into a single list: strategic, operational, discretionary, tail. That list does not hold up, because it answers two different questions at once.
The first question is what the money is for. The second is how the purchase was made. Tail spend belongs to the second question. A £40 taxi and a £40 seat on a design tool are both tail spend, but one is travel and one is software. Filing them together as though tail spend were a category of expenditure makes the whole taxonomy collapse, and it hides the thing that actually matters.
Two axes work better. Where they cross is where the problems concentrate.
| Category | What it covers | Direction of travel |
|---|---|---|
| Strategic and core | Direct costs, cost of goods sold, contracted suppliers. The spend with a business case behind it. | Not measured separately |
| Operational | The day-to-day cost of keeping the business running. | Up 26%, everyday shopping up 33% |
| Software and subscriptions | SaaS, licences, cloud services, tooling. | Up 43% in the UK, 37% in Italy |
| Professional services | Consultants, legal advice, ESG reporting, technical projects. | Up 59% in the UK, 41% in Italy |
| Travel and expenses | Travel, subsistence, mileage, client entertainment. | Up 12% |
| Discretionary | Valuable but not essential. First to be cut when budgets tighten. | Not measured separately |
| People and welfare | Benefits, welfare, training, equipment. | Not measured separately |
Axis Two: How it gets bought
-
Managed spend. Under contract, sourced and approved through procurement. A small number of suppliers accounting for most of the value.
-
Tail spend. Low-value, high-volume purchases spread across many suppliers. Individually trivial, collectively material.
-
The tail of the tail. One-off spot buys from suppliers who never get a supplier record at all.
-
Maverick spend. Bought outside the agreed route. Usually a symptom of a process that was too slow, not of a rule-breaker.
-
Invisible spend. Paid for personally and never claimed, so it never enters any system at all.
Invisible spend
41% of UK employees buy something for work and never claim the money back, and 15% only bother claiming for significant amounts. That spending is real, it benefited the business, and it appears in no category, no report and no supplier record. Any spend analysis built on the books alone is therefore wrong by a margin nobody can measure. This is a data integrity problem as much as an employee welfare one.
Why the classification matters
The reason both axes matter is that the right control depends on where a purchase sits on each of them.
Contracted strategic spend rewards sourcing effort. Negotiation, consolidation and supplier management all pay for themselves at that end, because the values are large and the supplier count is small enough to manage.
Tail spend rewards none of that. There is no negotiating leverage in a single £60 purchase, and the sourcing effort costs more than any available saving. What tail spend needs is a payment method with the policy already built into it, so the purchase can happen without an approval cycle and still stay compliant.
This is why applying purchase order processes to tail spend is self-defeating. It is what creates the £50 to £1,000 per order problem in the first place. The answer is not a faster purchase order. It is not raising one.
| Managed spend | Tail spend | Invisible spend | |
|---|---|---|---|
| Typical value | High per transaction | Low per transaction | Low per transaction |
| Supplier count | Small and known | Large and changing | Unrecorded |
| Control that works | Sourcing, contracts, negotiation | Pre-authorised cards with policy built in | Remove the need to pay personally |
| Control that fails | None. This is where purchase orders earn their keep. | Purchase orders and approval chains | Chasing claims after the fact |
| Visibility today | Usually good | Usually partial | None |
How to manage business spending
Four things to get right: govern the spending, pay for it deliberately, review it continuously, and measure whether any of it is working.
Govern: set the rules before the money moves
Spend governance is the difference between a proactive and a reactive finance function. A reactive process checks spending after it has happened and corrects what it can. A proactive process makes non-compliant spending difficult to do in the first place.
The practical distinction is where the policy lives. A policy in a PDF is a document employees skim once during onboarding and then guess at. A policy encoded into the payment method is a rule that applies itself at the point of purchase.
Only 39% of UK employees say they know exactly what can be expensed. That is not a problem a longer policy document solves. It is an argument for moving the rules out of the document and into the card.
Review: make month-end a non-event
Month-end is where a weak spend process presents its bill. Receipts get chased, codings get guessed, and the close takes days it should not take.
Almost all of that work exists because the data was captured late. A receipt photographed at the till, coded by the person who made the purchase, and matched automatically against a real card transaction needs no chasing. A receipt that surfaces in an envelope three weeks later needs a conversation, and possibly an argument.
Two things make the difference. The first is capture at the point of purchase rather than the point of claim. The second is an immutable record, so that every transaction carries a cardholder, an approval, a receipt and a category permanently, rather than having them reconstructed at the end of the month.
What month-end looks like when spending is captured at source
-
Spend happens. An employee pays on a company card. The transaction appears in the dashboard immediately, not at the end of a statement period.
-
Capture at source. The receipt is photographed in the app, then coded and tagged by the person who made the purchase and knows what it was for.
-
Automatic matching. The receipt is matched to the transaction. Anything missing is flagged while people still remember the purchase.
-
Continuous review. Finance reviews exceptions as they arise rather than working through a month’s worth in one sitting.
-
Export and reconcile. Categorised transactions sync to the accounting system or ERP with VAT treatment already applied.
-
Close. There is no backlog to clear, because nothing was left outstanding in the first place.
Measure: the numbers that tell you it is working
Spend management is easy to start and easy to leave half-finished. These are the measures that show which of the two has happened.
Most of them are available from day one, and almost nobody tracks them. The first two are the most revealing, because they quantify the spending that never made it into the books at all.
Spend management metrics worth tracking
-
Total reimbursed to personal accounts
-
Estimated unclaimed spend
-
Claims submitted late
-
Claims rejected, and why
-
?rel=0Policy exceptions per month
-
Working days to close the books
-
Cost per transaction processed
-
Purchase orders raised under £500
-
Total active suppliers
-
Suppliers with a single transaction
-
Duplicate subscriptions identified
-
Share of spend visible in real time
Spend Management: Compliance, records and auditing
What the record needs to show, what UK and European regulation now expects, and why forged receipts are a card problem rather than a checking problem.
What good record-keeping looks like
A defensible spend record ties four things together for every transaction: who spent the money, what they bought, what evidence exists, and who authorised it. If any one of those is reconstructed after the event, the record is weaker than it looks, and it is weakest at exactly the moment somebody asks to see it.
The direction of regulation across the UK and Europe is towards precisely this. Transaction-level digital records, produced at the time of the transaction, retained in a form that cannot be quietly amended afterwards.
United Kingdom: Making Tax Digital
Making Tax Digital is HMRC’s programme for moving tax reporting onto digital records and software submissions. It has arrived in stages since 2019, and the scope depends on what kind of business you are. This is where most summaries mislead.
Making Tax Digital for VAT already applies to every VAT-registered business. Digital records, and VAT returns filed through compatible software.
Making Tax Digital for Income Tax began on 6 April 2026, and it applies to sole traders and landlords rather than to companies. The first phase covers those with qualifying income above £50,000. That threshold falls to £30,000 from April 2027 and to £20,000 from April 2028. Businesses in scope keep digital records and send HMRC quarterly updates through compatible software instead of filing one annual Self Assessment return. Partnerships are not yet in scope.
So if you run a limited company, Making Tax Digital for Income Tax does not apply to you, and any guide implying otherwise is worth distrusting on its other points too. What does apply is the expectation underneath it: digital records, captured as you go. That expectation is not going to loosen.
United Kingdom: HMRC records and VAT evidence
Reclaiming VAT on business spending depends on holding valid evidence for the purchase. This is where card spending without captured receipts quietly costs money: if the receipt is missing, the VAT is generally not recoverable, and the loss never appears as a line item anywhere. It is simply absent from what you reclaim.
The same records support mileage and subsistence claims, which carry their own rules on what qualifies and at what rate. Capturing the evidence at the point of purchase is what makes all of it routine rather than a quarterly reconstruction exercise.
Italy. Italy has moved further than most markets. Electronic invoicing is long established, and Legge di Bilancio has mandated digital audit trails for transactions. For Italian businesses, traceability is not a future consideration. It is the current baseline.
France and the wider EU
France’s business-to-business electronic invoicing mandate takes effect on 1 September 2026. From that date every company must be able to receive electronic invoices, and large and mid-sized companies must also issue them and report the transaction data to the tax authorities. Small and micro-enterprises follow on 1 September 2027.
Invoices have to travel through certified platforms rather than being sent directly, in structured formats meeting the European EN 16931 standard. It is a move from auditing VAT after the fact to validating it continuously.
This is not a French peculiarity. It is one national implementation of the direction the whole EU is taking under the VAT in the Digital Age reforms, with intra-EU digital reporting to follow. Any business trading across European markets should expect structured, platform-mediated reporting to become the norm rather than the exception.
Dates in this area have moved before. Confirm the current position against DGFiP guidance before relying on it for a compliance decision.
Fraud, forgery and the limits of checking
74% of UK finance leaders worry about AI-generated fake receipts, against 41% in Italy. The concern is well founded, because a convincing fabricated receipt is now cheap to produce and hard to spot by eye.
What makes this manageable is not better checking. It is removing the document from its role as the primary evidence. When an employee pays on a company card, the transaction record comes from the card network rather than from the employee. A receipt captured at the till attaches to a payment that demonstrably happened, at a merchant that demonstrably exists, for an amount that matches. A forged receipt submitted as a reimbursement claim has no such transaction to attach to.
This is the part worth being clear about: reimbursement-based expenses are the vulnerable model, because the claim is the only record of the purchase. Card-based spending inverts that relationship entirely.
74%
Worry about AI-generated fake receipts
41%
Share the same concern
35%
Say employees forget to claim
36%
Name finance as the top source of ambiguous claims
Choosing a spend management solution
Audit your own organisation first, understand what you are really choosing between, then work out what it costs and how you roll it out.
Start with your own organisation, not a shortlist
Most software evaluations start by comparing vendors, which is the wrong end of the problem. Comparing features tells you which product is richer. It does not tell you which problem you have, and it does not stop you buying a solution to somebody else’s.
Answer these first. If you cannot answer several of them, that is itself the finding.
Ten questions to answer before you look at any vendor
-
How many people hold a company payment card today?
-
How many need one but do not have one?
-
How much did you reimburse to personal bank accounts last quarter?
-
How many working days does it take to close the books?
-
How many active suppliers do you have, and how many had a single transaction?
-
How many purchase orders did you raise last quarter for less than £500?
-
Can you see this week’s spending, or only last month’s?
-
Who interprets the expense policy today when somebody is unsure?
-
What share of receipts arrive within 48 hours of the purchase?
-
What happens today when somebody needs to buy something urgently?
| Option | When control applies | Who can hold one | Employee pays personally | Reconciliation |
|---|---|---|---|---|
| Reimbursement and spreadsheets | After the spend | Everyone, by default | Always | Manual |
| Corporate credit card | After the spend | A few staff, after credit checks | Often, when no cardholder is free | Manual, statement-led |
| Company debit card | After the spend | Whoever holds the card | Sometimes | Manual |
| Procurement card (P-card) | After the spend | Selected buyers | Sometimes | Partly automated |
| Expense claim software | After the spend | Everyone submits, nobody spends | Always | Claim processing only |
| ERP procurement module | Before the spend, via purchase orders | Requisitioners | Rarely | For PO-backed spend only |
| Spend management platform | Before the spend, at the card | Everyone who needs one | No | Automated |
-
Payment breadth. Physical cards, recurring virtual cards, single-use virtual cards and bank transfers, so the instrument can fit the purchase.
-
Controls that apply first. Limits by merchant, category, time and transaction, enforced at the point of payment rather than checked afterwards.
-
An app people will use. Real-time balances, fund requests and receipt capture at the till. Adoption is the whole game.
-
Real-time dashboard. Every transaction as it happens, filterable by team, project, merchant and date.
-
Accounting and ERP sync. Native integrations with the finance stack, carrying categories and VAT treatment across.
-
Multi-entity and multi-currency. Separate entities, several currencies, and reporting that consolidates across both.
-
Issuance at scale. Cards issued, frozen, reassigned and cancelled without waiting for third-party approval.
-
Audit-ready records. An immutable trail tying cardholder, approval, receipt and transaction together.
Regulatory status and where your money sits
This gets skipped in most evaluations and it should not. If a provider is going to hold your funds, ask under what permissions, in which entity, and what happens to the money if the provider fails.
Electronic money institutions are required to safeguard customer funds separately from their own, which is a materially different arrangement from an unregulated intermediary holding a balance on your behalf. Ask for the regulatory permissions in writing, along with the safeguarding arrangement, the security certifications, the data processing terms and where data is held.
A finance team that cannot answer these questions about its own supplier has a governance gap of exactly the kind this guide is about.
Working out what it actually costs
Headline pricing rarely decides the outcome. What matters is the total of the licence cost, the transaction costs, and the cost of the process you are replacing. The last of those is the one nobody calculates, and it is usually the largest.
Questions to put to a vendor on cost
-
Is pricing per user, per card or per entity?
-
What happens when a cardholder spends nothing that month?
-
What is the rate and margin on foreign currency spending?
-
Are there ATM or cash withdrawal fees?
-
What is the minimum contract term?
-
What does implementation cost, and is support included?
-
Which integrations are included and which are chargeable?
-
What does the process you are replacing cost in finance hours per month?
A realistic rollout
-
Week 1, Pick a pilot team. Choose one team with frequent, varied spending. Sales, operations or facilities usually works better than finance.
-
Weeks 1 to 2, Rewrite the policy as rules. Convert the written policy into limits, categories and approval thresholds. Anything you cannot express as a rule needs rethinking.
-
Week 2, Issue cards and wallets. Set up ringfenced wallets per team or project, then issue cards drawn against them.
-
Weeks 2 to 3, Onboard the pilot. Show people the app at the point of need. Adoption depends almost entirely on the first purchase going smoothly.
-
Week 4, Card first, reimbursement last. Make the card the default route and reimbursement the exception. This is the change that removes out-of-pocket spending.
-
Month 2, Connect the finance stack. Switch on the accounting or ERP integration and confirm categories and VAT treatment map correctly.
-
Months 2 to 3, Roll out and retire. Extend to the remaining teams, then withdraw petty cash and cut the shared credit card back to genuine exceptions.
-
Ongoing, Measure. Track the metrics from Chapter 3 and review policy exceptions monthly.
Proactive spend management in practice
What changes day-to-day once the rules are embedded in the payment method rather than in a document?
Firstly, Spending organises itself. Every transaction arrives already tagged to a team, project or cost centre, because the card it was made on belongs to one. Nobody allocates anything afterwards. The categorisation is a property of how the money was spent rather than a job somebody does at month-end.
Procurement enabled, not bypassed
An app that works at the point of need
Employees can see their balance, request funds and get them approved while they are standing in the shop. The alternative is a phone call to finance, or paying for it themselves. Requests that take minutes rather than days are the difference between a policy people follow and a policy people work around.
Petty cash, retired
A low-limit physical card does everything a cash tin does, with a transaction record attached to every purchase. No counting, no reconciling a float, no paper receipts in an envelope, and no untraceable spending.
Subscriptions and recurring payments under control
Issue a dedicated virtual card per supplier and every renewal becomes traceable to one card. Duplicate licences across teams become obvious rather than invisible, and cancelling a service means freezing a card rather than hunting through a statement. This matters more each year: UK software spend rose 43% in the first half of 2025.
Travel and expenses without the claim form
Cards carry the travel policy with them, so limits and blocked categories apply while the employee is booking rather than when finance reviews the claim. Receipts are captured at the till. There is no claim to submit, which means there is no claim to reject.
Departmental, multi-entity and international spend
Wallets keep each department, project or entity funded separately, so one team cannot spend another’s budget. Multiple currencies are held and spent without a personal card and a guess at the exchange rate, and reporting consolidates across entities and currencies together.
No more out-of-pocket spending
This is the change employees notice. 41% of UK employees currently buy something for work and never claim the money back, and 38% avoid spending altogether even where it would have helped the business. Giving people controlled access to company money removes both problems at once.
A workable alternative to the corporate credit card
Corporate credit cards have real advantages, but you cannot issue one to everybody. Credit checks, cardholder limits and slow issuance mean they concentrate in a few hands, which is exactly what pushes everyone else towards personal cards. See the comparison in Chapter 5 for the full picture.
Connected to your accounting software
Transactions, categories, receipts and VAT treatment sync into the accounting system or ERP directly. The value is not the time saved on data entry, though that is real. It is that the accounts and the spend data stop being two versions of the same month.
Procurement enabled, not bypassed
There is a version of this that procurement teams are right to be wary of. Handing every employee a card sounds like decentralised buying with the controls taken out.
Done properly it is the opposite. The controls move from the approval queue into the card itself, which means procurement stops processing low-value requisitions and starts working on the spend where sourcing effort actually pays. Onboarding a new supplier can take up to six months. Processing a single purchase order can cost between £50 and £1,000. Neither is a sensible use of procurement’s time on a £70 purchase.
Three shifts follow. Cut the purchase order pile, so small buys stop consuming the process built for large ones. Keep work moving, so teams stop inventing workarounds that nobody can see. Then use the transaction data to negotiate from a stronger position, because consolidated spend data across every team is leverage that fragmented spending never produces.
Leakage and fraud, prevented rather than detected
Controls that operate before a payment catch things retrospective review cannot. Blocked merchant categories stop out-of-policy purchases at the till. Per-transaction and time-based caps stop overshoots rather than reporting them. Real-time alerts surface unusual activity while it can still be stopped, and a card can be frozen instantly rather than through a bank’s cancellation process.
Card sprawl shrinks as well, because single-use virtual cards remove the reason to keep dormant cards active for occasional purchases.
What this changes, by role
Finance director
You see spending as it happens instead of reconstructing it monthly. Budgets are enforced rather than reported on. The close stops being an event. The bigger shift is that the finance team moves from processing claims to advising on the spending itself, which is the job you hired them for.
Procurement lead
Low-value buying stops arriving as requisitions. Suppliers can be paid without a six-month onboarding cycle. Tail spend becomes visible alongside your ERP data, so you can consolidate suppliers and spot duplicate spending. Your time goes to the contracts where negotiation actually moves the number.
Budget holder
Your budget sits in a wallet you can see, drawn down in real time. You approve requests from your team in the app rather than by email. You find out about budget burn while there is still time to do something about it, not in a variance report after quarter end.
Employee
You stop paying for work with your own money. You know what you are allowed to spend because the card tells you. You photograph the receipt at the till and then forget about it. There is no claim form, no waiting for reimbursement, and no wondering whether a small purchase is worth the hassle.
Conclusions: Where spend management is heading
Three things are converging, and all three point the same way.
Real-time compliance is becoming the default. Digital, transaction-level audit trails are already mandatory in Italy, and the direction of travel across the UK and the EU is the same. Businesses that capture spend data at the point of purchase will find themselves compliant by construction. Businesses that reconstruct records at month-end will be doing the work twice, once for the accounts and once for the auditor.
AI is cutting both ways. It has made receipt forgery cheap, which is why 74% of UK finance leaders now worry about it. It has also become a procurement responsibility in its own right: AI investment rose 130% among large businesses in the first half of 2025, and somebody has to source those tools and assess them for cost, capability and risk.
The control point keeps moving earlier. The long-run trend is away from checking spending after it happens and towards making non-compliant spending difficult to execute at all. Policy stops being a document that describes what should happen and becomes a property of the payment method itself. Everything in this guide is a version of that one shift.
The terms used in this guide, defined.
-
Spend management. The practice of controlling all the money a business pays to third parties, covering policy, approval, payment, evidence and reconciliation.
-
Expense management. The management of costs incurred by employees on the business’s behalf, and the process by which those costs are claimed and reimbursed. A subset of spend management.
-
Procurement. The function responsible for sourcing suppliers, negotiating contracts and managing supplier relationships.
-
Accounts payable. The function responsible for receiving, processing and settling supplier invoices.
-
Tail spend. The long tail of low-value, high-volume purchases spread across a large number of suppliers. Individually small, collectively significant, and usually the least visible spending a business does.
-
The tail of the tail. One-off purchases from suppliers a business buys from only once, who often never get a supplier record at all.
-
Maverick spend. Spending made outside the agreed purchasing route. Usually a symptom of a purchasing process that was too slow rather than deliberate rule-breaking.
-
Invisible spend. Money spent on the business’s behalf that never enters any system, typically because an employee paid personally and did not claim it back.
-
Spend under management. The proportion of total third-party spending that is actively governed by a defined process, rather than happening outside one.
-
?rel=0Purchase order (PO). A document authorising a purchase from a supplier before it happens, specifying what is being bought and on what terms.
-
Blanket purchase order. A single purchase order covering repeated purchases from one supplier over a period, rather than one order per purchase.
-
Three-way match. A control that checks the purchase order, the goods received note and the supplier invoice against one another before payment is released.
-
Procurement card (P-card). A payment card issued to selected buyers for low-value purchases. Reduces purchase order volume, but its controls and visibility typically apply only after the money has been spent.
-
Virtual card. A card that exists only as a number, issued instantly and usable online. Can be tied to a single supplier or a single purchase.
-
Single-use card. A virtual card that expires after one transaction, used for one-off purchases without onboarding a new supplier.
-
Ringfenced funds. Money allocated to a specific team, project or cost centre and unavailable for anything else, so budgets cannot be spent by the wrong people.
-
Wallet. A named balance within a spend management account, used to hold ringfenced funds for a team, project, entity or cost centre.
-
Merchant category control. A restriction that permits or blocks spending based on the type of business being paid, using the merchant category assigned by the card network.
-
Cost centre. A defined part of a business to which costs are allocated for reporting and budgeting purposes.
-
Out-of-pocket expense. A cost an employee pays personally on the business’s behalf, expecting reimbursement.
-
Reimbursement. Repaying an employee for a cost they paid personally. The alternative to giving them controlled access to company money in the first place.
-
Reconciliation. Matching transactions to their supporting evidence and to the accounting records, so the two agree.
-
Month-end close. The process of finalising the accounting records for a period. The point at which weaknesses in a spend process become visible and expensive.
-
Audit trail. The record of who did what and when. A strong audit trail is created at the time of the transaction and cannot be amended afterwards.
-
Immutable record. A record that cannot be altered once created, which is what makes it useful as evidence.
-
Spend policy. The rules governing what may be bought, by whom, up to what value and through which route.
-
Delegated authority. The value up to which a named individual may approve spending without escalating it.
-
Making Tax Digital (MTD). HMRC’s programme moving UK tax reporting onto digital record-keeping and software submissions. MTD for VAT applies to all VAT-registered businesses. MTD for Income Tax began in April 2026 for sole traders and landlords above an income threshold, and does not apply to limited companies.
-
Electronic invoicing. Issuing and exchanging invoices in a structured digital format that software can process automatically. Long mandatory between businesses in Italy, and mandatory in France for large and mid-sized companies from 1 September 2026.
-
Electronic money institution. A regulated firm authorised to issue electronic money and required to safeguard customer funds separately from its own.
-
Safeguarding. The regulatory requirement to hold customer funds separately from a firm’s own money, so those funds are protected if the firm fails.
-
Interchange. The fee paid between banks when a card payment is processed. Relevant when comparing the true cost of card-based spending.
-
Subsistence. Food, drink and accommodation costs incurred while working away from a normal place of work.
-
Mileage allowance. A per-mile rate paid to employees using their own vehicle for business travel.
Browse by Topic
Frequently asked questions about Spend Management
Spend management is the practice of controlling all the money a business pays to third parties, from contracted supplier costs to everyday employee purchases. It covers setting expense policies, issuing payment methods, tracking transactions in real time, and reconciling them with accounts. It is broader than expense management, which handles only employee claims.
Procurement is concerned with how a business buys: sourcing suppliers, negotiating contracts and managing supplier relationships. Spend management is concerned with controlling money at every point it leaves the business, whether or not procurement was involved. Procurement sits inside spend management, and most spend management problems involve purchases that procurement never saw.
Maverick spend is any purchase made outside the agreed buying route, for example paying a supplier on a personal card instead of going through procurement. It is usually a symptom rather than a cause: people work around a process because the process was too slow for what they needed.
Processing a single purchase order can cost between £50 and £1,000 on average, including review time. For low-value purchases, this frequently exceeds the value of the item being bought, which is why applying purchase order processes to tail spend tends to cost more than it saves.
Reclaiming VAT depends on holding valid evidence for the purchase, not on which payment method was used. Card spending without a captured receipt generally means the VAT cannot be recovered, and the loss never appears as a line item anywhere. Capturing receipts at the point of purchase is what makes the reclaim routine.
Making Tax Digital for VAT already requires every VAT-registered business to keep digital records and file VAT returns through compatible software. Making Tax Digital for Income Tax began on 6 April 2026 for sole traders and landlords with qualifying income above £50,000, falling to £30,000 in April 2027. It does not apply to limited companies. For spending, the effect either way is that digital, transaction-level records become the reporting format rather than good practice.
At minimum: what can be bought, by whom, up to what value, through which route, and what evidence is required. The more useful test is whether each rule can be expressed as a control on a payment method. Anything that cannot usually needs rewriting, because a rule that only exists in a document is a rule people will guess at.
Expense management deals with costs employees incur on the business’s behalf and the process for claiming them back. Spend management covers all third-party spending, including supplier invoices, contracted costs and procurement, as well as employee expenses. Expense management is part of spend management, not a separate thing.
Tail spend is the long tail of low-value, high-volume purchases a business makes across a large number of suppliers. Each purchase is small enough to seem unimportant, but together they account for a significant share of transactions and most of the administrative burden. It is usually the least visible spending a business does.
A procurement card is a physical payment card issued to selected buyers, with controls and reporting that generally apply after the money has been spent. A virtual card exists only as a number, can be issued instantly, and can be restricted to one supplier, one category or one transaction, so the controls apply before the payment goes through.
They should not have to, but many still do. 41% of UK employees buy something for work and never claim the money back, and 38% avoid spending altogether, even where it would have helped the business. Giving employees controlled access to company money through a company card removes the need for personal spending entirely.
You need to be able to show what was bought, when, by whom, for what business purpose, and with evidence of the amount paid. In practice this means retaining receipts or invoices alongside a record of the transaction. Records created at the time of the purchase are considerably stronger than records reconstructed later.
It depends on what you need. Corporate credit cards offer credit and are well-suited to a small number of senior cardholders, but credit checks and slow issuance mean you cannot practically give one to everybody. Prepaid company cards can be issued to anyone, with limits and category controls applied up front, which allows spending to be devolved without losing control.
A pilot team can typically be live within a fortnight, because cards can be issued immediately and the initial work is converting the written policy into rules. A full rollout, including accounting integration and the withdrawal of petty cash and legacy cards, usually takes two to three months.
Browse by category
Latest blog posts
Discover stories, tips, and resources to inspire your next big idea.






