Purchasing Card Program Evaluation for Procurement Teams

Procurement should evaluate a purchasing card program first on how well it enforces procurement policy at the point of sale and how good the transaction data is that comes back. Cash flow and rebate come second, because both depend on the program being used correctly.

That order runs against how most P-card programs get chosen. Finance compares rebate offers, picks one, and hands procurement a card that can buy from merchants the policy bans, returns a merchant name and a dollar amount, and leaves the program administrator chasing receipts. A first-quarter audit often finds the merchant blocks working fine. The trouble is the transactions with no receipt or approval attached, where nobody can say whether those purchases were allowed.

Key Takeaways

• Judge a P-card program on control design and data quality before rebate, because rebate only follows spend that stays on the card.

• Line-item Level 3 data lets procurement analyze category spend and contract compliance, and it lowers the supplier's cost on qualifying transactions.

• Set merchant category blocks, transaction limits, and monthly limits by role, and route approvals through the same rules as your procurement policy.

• Documentation, not fraud, is the measured weak spot in P-card programs, so test receipt capture and approval evidence during the evaluation.

• Purchasing cards can improve cash flow through billing-cycle float, but the gain is modest and paying the statement late can cost more than it earns.

What should procurement look for in a purchasing card program?

Procurement should look for a program that enforces its purchasing policy automatically, returns data it can use, and fits its approval and PO processes. Test each of those directly rather than taking a feature list on trust.

If you need the basics first, start with what a corporate purchase card is and how it changes the buying process. This page assumes you're already comparing programs. And if your evaluation is really about travel and expense cards, the criteria shift, and the guide to evaluating corporate card providers is the better starting point.

GFOA's best practice on purchasing cards, approved in September 2020, makes the same point from the public-sector side. It recommends four program elements.

• A purchasing card policy that conforms to the entity's overall procurement policy.

• A procedures manual that also addresses fraud prevention.

• A training program, with a manual, that keeps cardholders accountable.

• A program design that deals with the shift in reconciliation responsibility.

A card program that can't express your policy in its settings leaves procurement enforcing it by hand.

How much transaction data should a P-card return?

A P-card should return Level 3 data wherever suppliers send it, meaning line items with quantities, unit prices, product codes, and tax. Level 2 adds tax and a customer code to the basic merchant name and amount, and the basic record alone is close to useless for procurement analysis.

Line-item data is what turns P-card spend into procurement information. With it, procurement can do work a merchant name never supports.

• See which categories are moving to the card.

• Compare unit prices against contract pricing.

• Flag off-contract purchases by supplier and cardholder.

• Feed purchase price variance analysis with real numbers.

Without it, a $4,000 charge at an industrial distributor could be gloves, a pump, or a laptop. The differences between Level 2 and Level 3 card data decide how much of that detail you get.

Suppliers have a reason to send it, too. According to Visa's USA Interchange Reimbursement Fees schedule effective April 18, 2026, a commercial card-not-present transaction carries interchange of 2.70% plus $0.10, while a transaction that qualifies with Level 3 data carries 1.75% plus $0.10. A supplier that passes complete data pays noticeably less to accept your card, which makes the request easier to make.

Ask each program how it captures the data and how it reports the share of your spend arriving at each level. Then ask whether line items post to the ERP or stay in the bank's portal. Corpay's card programs post transaction data to 100+ ERPs, including NetSuite, Sage Intacct, Business Central, and Acumatica, so coded P-card spend lands in the same ledger procurement reports from.

What happens when suppliers don't pass Level 3 data?

When suppliers don't pass Level 3 data, the transaction arrives with a merchant name and an amount, and someone has to code and explain it by hand. That's usually the cardholder, working from a receipt.

Don't let that become a reason to reject a program, because no program controls what a supplier's processor sends. Treat it as a measurable gap. Ask for a report of your top 50 P-card suppliers by spend with their current data level, then work the list with procurement's supplier contacts. Receipt capture on the cardholder's phone covers the rest, as long as the program attaches the receipt to the transaction rather than storing it somewhere else.

How should MCC blocking, limits and approvals be set up?

Set MCC blocks, single-transaction limits, and monthly limits by cardholder role, and route any purchase outside those limits through the same approvals your procurement policy uses. The card should enforce the policy so procurement doesn't have to.

Merchant category codes are the first layer. Start from a block list that covers obvious misuse, such as cash advances, gambling, and personal services, then open categories by role. A facilities technician needs hardware and maintenance suppliers. An office manager needs office supplies and software subscriptions. Neither needs airline tickets on a P-card. Merchant-level controls, blocking or allowing a specific supplier, help where an MCC is too broad, and card controls and spend policies covers the options most programs support.

Set the dollar values from your own procurement policy and approval thresholds.

The federal government offers a public reference point for single-purchase limits. The micro-purchase threshold in FAR 2.101, effective March 13, 2026, is $15,000, the level below which federal buyers can generally purchase without competition. Most companies set P-card single-transaction limits well below that, but the logic carries over. Tie the card limit to the point at which your policy starts requiring a quote or a PO.

Blocks and limits are the easy part, and they aren't where programs usually fail. GAO's 2017 review of government purchase cards (GAO-17-276) sampled 300 transactions from more than 17 million in fiscal 2014 and found little evidence of potential fraud, but 22% of transactions lacked complete approval documentation. That's the control gap to test in your evaluation.

How should approval routing work for P-card purchases?

Approval routing for P-card purchases should follow the same thresholds and approvers as your procurement policy. Pre-approval applies above a dollar limit or in restricted categories, and cycle review covers everything else.

Look for two things. First, can the program require pre-approval for specific categories or amounts, so a cardholder requests a temporary limit increase that a named approver signs off on? Second, does the cycle-end review route each cardholder's transactions to their manager with the receipt attached, and does it escalate when approvals stall? If the answer to either is "the cardholder emails a spreadsheet," you'll be living with the GAO finding.

Can a P-card program match purchases to purchase orders?

Some P-card programs can match card transactions to purchase orders, and procurement should require it for any spend above its PO threshold. The match links the card charge to the PO and the receiving record, so card-paid orders get the same three-way evidence as invoiced ones.

This matters most when procurement uses the card as a payment method for POs rather than as a replacement for them. The purchase requisition and PO still drive approval, and the card simply pays. Ask whether the program reads PO numbers from Level 3 data or from the cardholder, how it handles partial shipments, and whether it works with your purchase order software.

What reporting does procurement need from the program?

Procurement needs reports it can run without asking finance, covering spend by category, supplier, cardholder, and department, plus a policy-exception report showing declines, out-of-policy purchases, and missing receipts or approvals.

The exception report is the one that tells you whether the program is working. A program administrator should see, each cycle, which cardholders have missing documentation, which merchants triggered declines, and which purchases went through under a category that procurement wants on contract. Spend-by-supplier reports belong in your supplier consolidation work, because P-card spend is often the long tail that never shows up in contract reviews.

Reconciliation sits next to reporting. Finance owns it, but procurement feels it when corporate card reconciliation falls behind and cardholders get chased for receipts weeks after a purchase.

Can purchasing card programs improve cash flow?

Yes, purchasing card programs can improve cash flow, mostly through billing-cycle float. You pay the supplier at purchase, and you pay the card issuer when the statement comes due, often several weeks later.

The gain is real but modest, so do the arithmetic before you sell it internally. On illustrative numbers, a monthly billing cycle with 25 days to pay gives an average of about 40 days between purchase and payment. If those purchases would otherwise be paid on net 30 terms, the card adds about 10 days. On $10 million of annual P-card spend, that's around $274,000 of extra average cash, worth about $11,000 a year at a 4% return. That's useful, but it isn't the reason to run a P-card program.

The bigger cash effects are elsewhere. Administrative cost is one. GAO's 2004 testimony on purchase cards (GAO-04-717T) cited an Army Audit Agency study that found the card saved $92 per transaction compared with a purchase order. That figure is two decades old and from a government setting, but the mechanism still holds, because a card purchase skips the PO, the invoice, and the payment run. Payment cost is another. According to AFP's 2022 Payments Cost Benchmarking Survey, issuing a paper check costs a median $2 to $4.

Rebate is the third, and it depends on paying on time. The Texas Comptroller's July 2024 audit of the statewide Citibank charge card contract found agencies lost about $1,518,236 in fiscal 2023, half the minimum rebate, by paying statements late, and the contract pays nothing at 31 or more days. Corpay returns $800M+ in rebates per year to businesses on its programs, and the cash only shows up when the statement is paid within terms. A program that stretches payment for float can give back more in lost rebate than the float earns, which is why card programs and cash flow need to be planned together.

Scale isn't the constraint it once was. GSA SmartPay ran $39.4 billion in program spend in fiscal 2025, averaging $480 per transaction, according to its Statistics and Reports page, and the Nilson Report put US commercial card purchase volume at $2.138 trillion in 2023. Mid-market companies can get the same controls the large programs use, so ask for them.

Run purchasing cards inside one Corpay commercial card program

Corpay's purchasing cards give procurement role-based limits, merchant controls, and transaction data that posts to the ERP, with single-use virtual cards available for supplier payments that shouldn't sit on a plastic card. They sit inside one Corpay commercial card program, so P-card, virtual card, and AP spend count together toward the rebate tier.

Corpay connects to 100+ ERP systems, including NetSuite, Sage Intacct, Business Central, and Acumatica, and programs go live in weeks. If supplier payments are the bigger problem, Corpay's fully managed AP pays suppliers by virtual card, ACH, or check from one approval flow. For how the pieces fit, see the one card program approach.

Frequently Asked Questions

What is the best purchasing card for procurement teams?

The best purchasing card for procurement teams is one that enforces procurement policy at the point of sale and returns usable data. Look for role-based limits and merchant controls, Level 3 data, and approval routing tied to your policy, then test them in a pilot with real cardholders.

What are the risks of a P-card program?

The main risks of a P-card program are missing documentation, out-of-policy purchases, and misuse. GAO's 2017 review found little evidence of fraud but frequent documentation gaps, so receipt capture and approval evidence matter as much as merchant blocks and limits.

What should mid-market companies look for in a purchasing card?

Mid-market companies should look for the same controls large programs use, plus a credit line and support model sized for their spend. Ask for ERP posting of coded transactions, and pilot with real cardholders before rolling out.

How much float does a P-card give you?

A P-card on a monthly billing cycle with about 25 days to pay gives roughly 40 days on average between purchase and payment. Compared with net 30 invoice terms, that's about 10 extra days of cash, as long as the statement is paid on time.

Who should own the P-card program, procurement or finance?

Procurement should own the P-card program's policy, controls, and supplier strategy, and finance should own reconciliation, payment, and the issuer relationship. Shared ownership works when both teams sign off on limits, merchant blocks, and approval routing before cards are issued.

David Luther

Product Marketing Program Manager

David Luther, MBA is a product marketing program manager with years of experience in commercial banking, finance, and technology sectors, with research and writing appearing in financial publications.

Commercial Cards

Procure-to-Pay