Expense management dashboard: spending and exceptions

An expense management dashboard shows what employees spent, what approved reimbursements remain unpaid, and which claims carry recorded exceptions. Build those views from the same claim-level export, with a fixed reporting date and a clear payment status. A useful dashboard connects each total to claims a finance reviewer can inspect.

This is an employee expense view. Broader departmental budgets belong in cost center reporting; the dashboard here answers questions about categories, employee reimbursement timing, and unresolved documentation.

Start with the claim cohort and the reporting date

Use one row per claim line, with a unique line identifier. Retain the claim identifier too: a claim containing several expense lines must not become several claims in your exception count. Include category, reimbursable amount, currency, approval date, payment date, payment channel, and exception status.

Choose the cohort before calculating anything. The example below covers claims approved during September 2026, as observed on September 30, 2026. Every amount is USD, every claim has one line, and each payment settles its claim in full. Partial reimbursements would require a paid-amount column and an outstanding amount for each claim.

For a real export, preserve the original category and any cleaned category side by side. A category mapping should combine spelling differences without hiding whether a charge was travel, meals, or software. Keep the original exception reason as well as its open or resolved status.

A worked employee expense dashboard

A US distributor's finance team reviews this compact approved-claim export. Dates below are in September 2026. A dash means no payment recorded by the reporting date; it does not mean a zero-dollar claim.

The first panel summarizes the approved cohort, rather than treating cash paid as spending:

Travel is 3,600 / 4,380 = 82.2% of approved spending. That identifies the largest category, but the oldest unpaid claim is in meals. Category size and reimbursement delay point to different review questions. Percentages here are rounded to one decimal place; dollar totals are exact.

In a worksheet with category in column B and approved amount in column C, the travel total is:

=SUMIFS(C2:C7,B2:B7,"Travel")

Microsoft documents SUMIFS as adding amounts that meet specified criteria. Use the same category definitions for the paid and unpaid summaries. Covirage is not affiliated with Microsoft.

Age only the outstanding reimbursement balance

For this dashboard, age begins at approval. Subtract the approval date from September 30, 2026, for unpaid claims only. E102 is 15 days old, E103 is 10, E105 is 5, and E106 is 2.

Unpaid reimbursement balance = Approved reimbursable amount − Amount reimbursed by the reporting date

These are reporting buckets, not a claim about what payment terms should be. The finance owner should compare the oldest items with the company's own reimbursement timetable and check whether a documentation question explains the delay.

The proof is two reconciliations: 1,680 + 900 + 360 = $2,940, and 1,440 + 2,940 = $4,380. The first proves that every unpaid dollar sits in an aging bucket. The second proves that payments plus the remaining balance equal the approved cohort.

Show exceptions as counts and dollars

E102 and E105 have open exceptions. The count rate is 2 / 6 = 33.3% of claims. Their combined amount is 360 + 180 = $540, or 540 / 4,380 = 12.3% of approved dollars. Display both denominators: the count rate describes review volume, while the dollar share describes the amount attached to those claims.

Do not label the whole $540 as disallowed spending. The IRS accountable-plan guidance describes requirements including a business connection, adequate accounting, and returning excess reimbursements. An internal exception flag alone does not establish whether those requirements are met. The dashboard gives the responsible reviewer a list to investigate.

If one claim has several exception reasons, count that claim once in the overall exception rate. A separate reason table can contain several entries, but its row count will not reconcile to distinct claims without that distinction.

Change one input: record an additional reimbursement

Now change only E102's payment date from blank to September 30. Its $360 moves from unpaid to paid. Approved spending remains $4,380; paid claims become 1,440 + 360 = $1,800, and unpaid claims become 2,940 − 360 = $2,580.

The oldest aging bucket becomes zero, and 1,680 + 900 = $2,580 still reconciles. The exception count remains unchanged because the documentation flag was not resolved. A payment event and an exception-resolution event are different facts.

Where this dashboard goes wrong

• Mixing cohorts. September-approved spending cannot reconcile to all September payments if those payments include older claims.
• Calling unpaid claims overdue. Age measures elapsed time. Overdue status needs the applicable due date or payment policy.
• Counting company card charges as employee debt. Identify who was paid and who remains owed money.
• Counting exception reasons as claims. Use distinct claim identifiers for the headline rate.
• Using a later payment in an earlier snapshot. Exclude payments after the stated reporting date when rebuilding historical balances.

Finance operations owns the reconciliation; policy owners interpret exceptions. Put the resulting findings into management reporting, with the approved cohort and reporting date visible. The wider management accounts still need ledger-based expenses, while a Power BI expense dashboard addresses a separate dashboard-building approach.

Review your own expense export in Covirage

Upload an expense export containing claim identifiers, amounts, dates, and recorded exception flags; Covirage maps the columns, and its tools compute and check the category totals and reimbursement balances. The external AI model explains the results with every figure traced to its rows. Create a free Covirage account and bring a sample of your own export.

See cost variance analysis for related spending questions.