Corporate Card Guide

A corporate card is easy to describe in one sentence and genuinely hard to understand in practice, because most of what makes it different from the card in your pocket is invisible.

The card itself is plastic or a string of digits. What distinguishes a corporate card is the administrative layer behind it: who issued it, in whose name, under whose liability, with which rules attached, and where the transaction goes once it posts.

The cards hub answers "what does this phrase mean"; the comparisons hub publishes decision frameworks. This guide is the third kind of page — a sequential explainer assuming no prior vocabulary.

A corporate card is a payment card issued to a company rather than to an individual, and given to employees to spend on the company’s behalf. The account belongs to the business. The employee is a cardholder, not a borrower. That single distinction produces almost every other difference you will encounter.

Compare it to a consumer card, where one person applies, is assessed, owes the balance and decides what to buy. On a corporate card the company does the first three, and a policy — not the cardholder’s judgement alone — governs the fourth. The cardholder is spending someone else’s money with delegated authority.

Account holder

The legal entity the account belongs to — on a corporate card, the company.

Cardholder

The named person a card is assigned to. They hold spending authority, not the debt.

Liability

Who is legally responsible for repaying what was spent.

Program administrator

Whoever issues and closes cards, sets limits and handles exceptions.

Authorisation

The real-time yes-or-no decision on a single transaction.

The second thing to understand is that "corporate card" describes a model, not a settlement structure. Some corporate cards are charge cards settled in full every cycle; others allow a balance to be carried. The word tells you about ownership and administration, not about repayment. That is why the brex corporate card and brex corporate credit card pages exist separately: the phrases are used interchangeably in search but point at different attributes.

People searching for a brex corporate card are usually trying to establish those five attributes for a specific product. Only the provider can confirm them. What this guide can do is make sure you know which five questions to ask.

A card program is not something you switch on. It is a sequence of decisions, and the order matters because each one constrains the next. Companies that issue cards first and write policy later spend the following year retrofitting rules onto habits that have already formed.

1. Decide what the program is for

   Software, travel, ad spend, vendor payments, or all of it. The mix decides whether you need mostly physical or mostly virtual instruments.

2. Write the spending policy in plain language

   One page: what people may buy, what needs approval, what needs a receipt. If you cannot write it, you cannot configure it.

3. Define roles rather than people

   A few limit-and-category templates by role, so issuing to the next hire is a lookup rather than a debate. See employee cards.

4. Match instrument to spending pattern

   Physical for in-person spend, single-vendor virtual for subscriptions. The virtual card guide covers the choice.

5. Configure controls before first issuance

   A card should never exist unconfigured. See the spending controls guide.

6. Connect the data path to accounting

   Decide where transactions land and who codes them before volume arrives.

7. Set a review cadence

   A recurring entry covering dormant cards, unused limits and role changes. Without it, limits only go up.

Steps two and three are the most often skipped and the cheapest. Policy written before configuration means every later control has a stated reason, which makes exceptions easier to judge and easier to refuse. The employee spending page covers how policy reaches the people it applies to.

This is the single most common point of confusion in the subject, so it belongs in this guide as a summary — with the full framework kept where it belongs, on the corporate card vs credit card comparison.

The two phrases describe different axes. "Corporate card" is about who holds the account and administers it. "Credit card" is about how the balance behaves — whether it can be revolved, and therefore whether interest can accrue. A card can be both, either or neither: a corporate charge card is a corporate card that is not a credit card, and a personally guaranteed small-business credit card is the reverse.

If you remember one thing: ask "who is liable" and "is the balance settled in full" as two separate questions. Conflating them is what makes the terminology feel arbitrary. The business credit card guide takes the credit-product side, and the brex credit card page covers the phrasing.

Limits are where a card program earns trust or loses it. Set them too low and people put work expenses on personal cards, destroying the data you built the program to get. Set them too high and the card stops being a control. The useful framing is not "how much should this person be allowed to spend" but "what is the smallest ceiling that lets this role work without asking permission".

Limits come in layers, and a transaction has to pass every applicable one. Knowing which layer bit is the difference between a five-minute fix and an afternoon of confusion.

A method for setting the first set of limits

1. Look at what the role actually spent over recent months, from whatever records you have.
2. Set the per-period ceiling near the observed normal, not the observed maximum.
3. Set the per-transaction ceiling at the largest purchase the role should make alone.
4. Add category permissions matching the written policy, then remove ones nobody has needed.
5. Tell the cardholder both numbers and how to request a change. Silent limits generate tickets.
6. Revisit after one full cycle. First-pass limits are a hypothesis, not a decision.

None of this involves a specific amount, because the right amount is a function of your business and nobody outside it can tell you what it is. What generalises is the method. The card limits page goes deeper on layer interaction, and budgets covers the group-level ceiling.

• Every card has a named owner, and the owner is a person rather than a team inbox
• Card creation and closure sit inside the onboarding and offboarding checklists
• Receipt and coding requirements are identical across every instrument type
• Exceptions are logged with an expiry, so temporary increases actually expire
• Dormant cards are reviewed on a schedule and closed rather than left available
• Someone owns the monthly review, and it lives on their calendar

The downstream half of the program is expense management: turning each posted transaction into a documented, coded, reconciled record. That has its own guide — the corporate finance guide — and its own cluster at expense management.

From there the natural next reads are the corporate card page for the term-level treatment, the spending controls guide for the control layer, and the startup card guide if your company is early enough that card structure has to substitute for process.

A reminder: this is an independent editorial project. We hold no accounts, issue no cards and speak for no provider. Anything commercial comes from the provider directly.