Brex

Brex exposes several object types through its API: card transactions, expenses, budgets, users, departments, and spend limits. This integration focuses on three of them — card transactions, expenses, and budgets — which together cover the full arc of corporate spend from raw financial event to compliance record to budget accountability. Contacts, departments, and locations are available but serve organizational segmentation rather than individual performance measurement. Five KPIs were retained from a longlist of eleven candidates, selected against three criteria: ability to attribute to an owner, resistance to gaming, and balance between leading and lagging indicators.

Spend volume and budget accountability

Spend per Employee measures the total amount charged to corporate cards by each employee in the period. Budget Utilization Rate measures, for each budget owner, what proportion of the approved budget has been consumed relative to the limit. These two indicators address the same question — whether spending is proportionate — from different vantage points. Spend per Employee provides a bottom-up view of individual consumption without reference to what was approved. Budget Utilization Rate provides a top-down view of whether the owner is tracking toward their allocation.

Reading them together surfaces a diagnostic that neither provides alone. A high Spend per Employee paired with a low Budget Utilization Rate may indicate that spend is occurring outside of formal budget structures — either on budgets not yet assigned or through channels not captured by the budget framework. The reverse pattern, a high utilization rate with moderate individual spend, may signal that budget allocation was too conservative rather than that employees are overspending. The combination forces the question of whether the budget model is fit for the actual spending pattern.

Budget Utilization Rate carries a specific gaming risk: a budget owner aware of the metric can delay discretionary spend toward period end to appear under-budget mid-period. This manipulation is visible when mid-period utilization rates are consistently low and end-of-period spend accelerates sharply. Spend per Employee, which aggregates at the individual transaction level without a budget reference, is unaffected by this behavior and provides a corrective signal.

Compliance process discipline

Average Submission Lag measures the average number of days between when a transaction settles and when the corresponding expense is submitted by the employee. Receipt Attachment Rate measures the proportion of expenses submitted with a receipt attached. Policy Violation Rate measures the proportion of expenses that are flagged as non-compliant with company policy.

These three indicators describe adjacent but distinct dimensions of the expense compliance process. Submission Lag is a leading indicator: employees who submit expenses late create a backlog that delays accounting close, and the pattern is detectable before it becomes a period-end problem. Receipt Attachment Rate and Policy Violation Rate are lagging indicators that measure the quality of what gets submitted, not the timeliness. An employee who submits quickly but without receipts, or within policy on category but with systematic documentation gaps, would score well on lag and poorly on the other two. No single indicator captures the full compliance profile.

The tension between Submission Lag and Policy Violation Rate is particularly useful. An employee with a very low submission lag and a high violation rate may be submitting quickly but without due attention to policy — speed and compliance are not the same behavior. Conversely, a long lag combined with a low violation rate may reflect an employee who is thorough but slow. The ideal profile is low lag and low violation rate, and the combination of the two indicators is necessary to distinguish between the four possible behavioral quadrants.

Receipt Attachment Rate is the least gameable indicator in this block. Improving the rate requires genuinely attaching receipts; the only way to inflate the metric is to fabricate documentation, which is a separate category of risk. This makes it a reliable compliance proxy and an anchor for reading the other two indicators in context.

Scope and limits of the integration

Brex records settled card transactions and enriched expense records, but it does not measure the business justification or strategic appropriateness of spend. An employee with a low Policy Violation Rate and perfect receipt attachment may still be making spending decisions that are suboptimal from a procurement or vendor management standpoint; the API data reflects procedural compliance, not commercial judgment. Similarly, Spend per Employee reflects consumption but not productivity — high individual spend may be entirely appropriate in a sales role involving client entertainment and travel, or it may signal lack of discipline, and the metric alone cannot distinguish between the two contexts.

The integration also only captures activity routed through Brex corporate cards and the Brex expense workflow. Spend on personal cards pending reimbursement outside of Brex, vendor invoices, or direct wire transfers are not visible. The reliability of these KPIs depends directly on the degree to which corporate spending is channeled through the platform: organizations with high Brex adoption capture a representative signal; those where Brex covers only a fraction of overall spend will see a partial picture that may systematically misrepresent individual or team spending patterns.