Solutions cluster hub
Card and Spend Solutions by Company Type
The mechanics of a company card never change. Authorisation is authorisation, a merchant category code is a merchant category code, and a receipt still has to be attached to something. What changes — completely — is who decides, how many cards exist, how much process is tolerable and which part of the system breaks first. This hub describes four company contexts and routes you to the one that matches yours.
Information only. This site is a reference project. We do not provide account access, financial services, card applications, payments, credit decisions or official support, and we never ask for account or financial credentials.
Orientation
Why company type changes the answer
Two companies can run technically identical card programs and have completely different experiences of them. The difference is almost never the card. It is the ratio of spending decisions to people available to review them.
A card program is a way of pushing spending decisions outward while keeping the consequences visible. How far outward you can safely push, and how much visibility you need in return, is a function of company size, spending mix and finance headcount — not of which card is in the wallet.
This matters because most advice about company cards is written as if there were one correct configuration. There is not. A six-person company that issues a card per vendor and never holds a spending meeting is operating correctly. A hundred-and-forty-person company doing the same thing has a problem, because nobody owns the aggregate. The spending controls page describes the mechanisms; these four pages describe when each mechanism starts earning its keep.
This site is an independent editorial project. We are not Brex, not an issuer, not a broker and not affiliated with any card provider. There is no application here, no account area and no pricing. What follows is a structural description of how card and spend requirements shift across company contexts, written so that whatever provider documentation you read next becomes easier to interpret.
- Decision volume
- How many independent spending decisions happen per month. This grows roughly with headcount and vendor count, and it is the number that eventually forces structure.
- Review capacity
- How many of those decisions a human can actually examine. In early-stage companies this is a founder’s spare attention; later it is a named finance role.
- Attribution
- The ability to say, without investigation, which team or purpose a charge belongs to. Attribution problems appear long before control problems.
- Process weight
- How much approval and documentation ceremony sits between wanting to buy something and buying it. Too much and people route around it; too little and month end absorbs the cost.
Cluster map
The four contexts
Each page describes a company profile, the spending pattern that comes with it, and the card and control structure that fits without over-engineering. Start with the one that matches where you are now, then read the next one up to see what is coming.
Startups
Little or no finance headcount, spending concentrated in software, cloud infrastructure and advertising, and a founder who is also the approver. Card structure has to substitute for process that does not exist yet.
Read more Owner-operatedSmall Business
A handful of cardholders, a blurred boundary between personal and business spending, and an external bookkeeper who sees the transactions weeks later. Policy can be light; attribution still cannot be skipped.
Read more ScalingGrowing Businesses
The twenty-to-two-hundred band where informal arrangements stop working. Shared cards, unexplained charges and a week-long month end are the standard symptoms of an outgrown setup.
Read more FunctionFinance Teams
The controller’s view: policy enforcement, complete audit evidence, clean coding, a predictable close and no surprises. Written for the people who own the consequences of everyone else’s card.
Read moreThe four are not a strict sequence. Plenty of small businesses never become growing businesses and have no reason to want to. Plenty of startups acquire a finance function before they acquire a hundred employees. But the order of problems is fairly consistent: attribution breaks first, then visibility, then policy, then evidence. Reading the page one step ahead of your current size is usually more useful than reading the one you are in.
- Cards hub The eight card reference pages and the taxonomy that connects them — what each card term actually means.
- Spend management hub Controls, limits, budgets and employee spending: the mechanisms referenced throughout these four pages.
- Expense management hub What happens after authorisation — coding, documentation, reconciliation and close.
Variables
What actually changes between them
Four variables account for most of the difference between these contexts. Everything else — which instruments you issue, how strict the limits are, how often you review — follows from where you sit on these axes.
Read the table as a description of pressure rather than prescription. Nothing in the right-hand column is a rule; it is the failure mode that tends to arrive first when a company keeps the previous stage’s structure a little too long.
| Context | Who approves | Cards in issue | Process weight | What breaks first |
|---|---|---|---|---|
| Startup | A founder, informally, often after the fact | Few physical, many single-vendor virtual | Deliberately minimal | Attribution — nobody can say which experiment a charge belongs to |
| Small business | The owner, who also holds most cards | A small fixed set, rarely changing | Light, but documentation still required | Separation — personal and business spending blur on the same instrument |
| Growing business | Managers, then a matrix as headcount rises | One per employee plus vendor and team cards | Rising, and resented if unexplained | Visibility — managers cannot see their own team’s spend |
| Finance team view | Policy, enforced automatically, with named exceptions | The whole program, reviewed on a schedule | Calibrated against close deadlines | Evidence — items reach close without coding or receipts |
An editorial framing used across this site to organise the four solution pages. It describes common patterns, not any provider’s product tiers or capabilities.
Who approves
Approval starts as a person and ends as a rule. The transition point is roughly where the approver stops recognising every vendor name on the statement. After that, a single approver becomes a bottleneck and an approval matrix becomes worth its overhead. See employee spending.
How many cards
The useful metric is not cards per employee but cards per distinct spending purpose. That number climbs faster than headcount, because vendors accumulate. Virtual cards are what make a high card count administratively cheap rather than chaotic.
How much process
Process weight should track the cost of a mistake, not the size of the company. A five-person team can run almost no ceremony; a company with departmental owners needs budgets so that "we are over" is a fact rather than a discovery.
What breaks first
Attribution, visibility, policy, evidence — usually in that order. Each is fixed by a different mechanism, and applying the wrong one produces bureaucracy without relief. Expense controls covers the evidence end.
Common ground
What stays the same everywhere
Before reading a stage-specific page, it is worth being clear about the parts that do not vary. These hold in a three-person company and in a three-hundred-person one, and they are the reason the four pages can share a vocabulary at all.
Every card program, at every size, does the same five things: it issues a credential, attaches rules to it, evaluates those rules at authorisation, captures the transaction, and enriches that transaction until it can become an accounting entry. The Brex Card page sets out that sequence in detail and is the best single starting point if the vocabulary here is unfamiliar.
Relative spend by period — shape only, no real figures.
- Only limits and merchant category rules genuinely prevent spending; approvals and receipt rules document and shape it
- A card with no owner is a reconciliation problem waiting to be discovered at month end
- Authorisation data is thin — amount, merchant descriptor, timestamp, category code — and everything else is attached afterwards
- One instrument per spending purpose beats one instrument per person for anything recurring
- Offboarding a person and cancelling their card are the same event, whatever the size of the company
- Commercial terms — fees, rates, rewards, credit limits, eligibility — come from the provider, never from a reference site
The card-side reading for all four contexts is the same: corporate cards for the company-liability model, business credit cards for the credit and guarantee model, employee cards for delegated authority and virtual cards for purpose-scoped issuance. The control-side reading is card limits and spending controls. What differs between the four pages is emphasis and sequencing, not the underlying model.
Transitions
Where companies get stuck
The uncomfortable moments are the transitions, not the stages. A structure that was correct eighteen months ago becomes the source of the current complaint, and the complaint is rarely phrased as "our card structure is out of date".
-
From personal cards to company cards
The founder-funded phase ends when reimbursements become a scheduling problem and nobody can separate personal from company spend. This is the first structural decision, and solutions for startups treats it directly.
-
From one card to many
Sharing one credential across a team feels efficient until the first disputed charge. The fix is instrument-per-purpose issuance, not stricter rules on the shared card.
-
From founder approval to manager approval
Once the approver no longer recognises the vendors, approval has to be delegated. Delegation without visibility is worse than no delegation, which is why budgets and limits arrive together. See growing businesses.
-
From manual coding to rules
Coding every transaction by hand is survivable at a few dozen items a month and not at a few hundred. Expense automation covers what can genuinely be automated.
-
From "the numbers" to audit-grade evidence
At some point someone external needs to be able to reconstruct a decision from the record alone. That standard is higher than internal comfort, and it is the subject of solutions for finance teams.
None of these transitions requires changing provider, and none of them is solved by a product feature alone. They are governance changes that a card program either supports or obstructs. The corporate card guide and the corporate finance guide work through the governance side at length.
Reading order
Where to start
Most readers need two pages from this cluster and one from elsewhere on the site. This is the shortest path through.
If you are setting a program up
- Read the page for your context: startups, small business, growing businesses or finance teams
- Then Brex Card for the shared vocabulary
- Then card limits to size ceilings sensibly
- Then the startup card guide if you have no finance function yet
If you are fixing one that is failing
- Identify which of the four failure modes you actually have before changing anything
- For attribution, go to virtual cards
- For visibility, go to budgets
- For policy, go to spending controls and the spending controls guide
- For evidence, go to business expenses
FAQ
Frequently asked questions
How do I know which of the four pages applies to my company?
Use finance headcount rather than employee headcount as the first test. If nobody spends most of their week on finance, read startups or small business depending on whether your spending is software-heavy or operations-heavy.
If you have at least one dedicated finance person and managers who own budgets, read growing businesses. If you are the finance function, finance teams is written from your side of the desk.
Do smaller companies need fewer controls or different ones?
Different ones, mostly. Small companies can run with very little approval ceremony because the approver knows every vendor personally, but they still need clean attribution — otherwise the bookkeeping cost simply moves to month end. The usual small-company answer is heavy use of purpose-scoped virtual cards and very light approval process.
What is the first sign that a card setup has been outgrown?
Someone asking "what is this charge?" about a transaction on a card they administer. That single question means attribution has broken, and it reliably appears before any visible control failure. The second sign is month end taking materially longer than it did two quarters ago.
Does adding limits and approvals slow a company down?
Badly designed ones do. Controls that are tighter than the work requires generate exception requests, and exception requests are pure overhead for both the requester and the approver. Worse, they push people back onto personal cards, which reintroduces exactly the attribution problem the program was meant to solve.
The design goal is the smallest amount of spending authority each role needs to work without friction, revisited on a schedule. Card limits covers how to size that.
Is a corporate card structure appropriate for a very small company?
Sometimes, and the honest answer depends on liability preferences rather than size. Company-liability corporate cards and personally guaranteed business credit cards are different structures with different consequences if the company cannot pay. Corporate card vs business card sets the two side by side without recommending either.
Do these pages tell me which provider to choose?
No. They describe requirements by company context so that you can evaluate any provider against your own situation. We publish no ratings, no rankings and no commercial terms. Brex vs other corporate card solutions offers an evaluation framework rather than a verdict.
Is this site connected to Brex in any way?
No. This is an independent, non-commercial informational project. It is not affiliated with, endorsed by, sponsored by or operated by Brex, and it does not speak for the company. There is no login, no account area, no application form and no request for card, banking or identity information anywhere on the site.
Sources and reference basis
- Practice Widely documented corporate card administration patterns: role-based issuance templates, budget ownership, delegated approval and periodic access review.
- Reference General payment-network reference material describing authorisation, merchant category classification and the data available at the point of a card transaction.
- Reference Standard accounting and internal-control literature on segregation of duties, documentation thresholds and period-end cut-off.
- Method Our methodology and fact-checking policy describe how these pages are researched, written and corrected.