Your 5% pay-rise budget needs three totals, not one

The 5% figure is a test input, not a pay recommendation

The fictional employer reports in US dollars and reviews three annual base salaries: $100,000, €60,000 and £50,000. Each receives a hypothetical 5% increase effective 1 July in a calendar-year budget.

Assume six equal monthly salary periods remain, with no headcount or working-time changes. Planning rates are invented: €1 = $1.10 and £1 = $1.25. They are not current exchange rates. The exercise excludes employer taxes, benefits, bonus effects and payroll-specific rounding.

A $11,425 annual commitment can cost $5,712.50 in this year

The annualised increase is the ongoing base-pay commitment. The six-month figure is this year’s simplified incremental base-pay cost. Neither includes the existing salary bill.

Now move only the pound-denominated increase to 1 October. Its three-month impact becomes $781.25, and the current-year total falls to $4,931.25. The annualised commitment remains $11,425. The effective date changed; the approved pay rise did not.

Do not make a manager absorb currency movement as if it were merit

Return to the euro salary alone. Before the increase, €60,000 at the fictional $1.10 rate is $66,000. After a 5% increase, €63,000 at the same rate is $69,300. The constant-rate annual increase is $3,300.

Suppose Finance instead forecasts the new salary at a fictional $1.20 per euro. It becomes $75,600, which is $9,600 above the old $66,000 translated salary. Calling that entire difference a pay-rise overspend would misdescribe the decision.

Other consistent bridge conventions can allocate the interaction differently. Label the convention. Keep the original $3,300 constant-rate cycle decision visible alongside the updated forecast; they answer different questions.

Pave: identify who can change cycle exchange rates

Pave’s currency configuration guide says cycle rates are frozen at the start rather than automatically following market changes. It also documents CSV rate updates before or during planning, with upload disabled after finalisation, and a change-history log.

That makes “frozen” more precise than “nobody can change it.” Ask a finalist to identify the active rate set, permitted changes and how a change affects already-approved budgets. Show the manager view and Finance export using the same scenario.

This is a documented product boundary, not a claim that we have operated a Pave cycle. Require the actual contracted configuration to reproduce your approved method.

Lattice: distinguish annual salaries from hourly rates

Lattice’s compensation-field guide requires base pay, currency, effective date and pay type. For salaried employees it expects an annual rate; hourly employees use an hourly rate.

Use that distinction in every demo. A monthly salary accidentally loaded as annual pay can produce a perfectly calculated percentage on the wrong base. An hourly employee also needs an agreed working-hours assumption before you compare annual cost.

Do not fix the discrepancy with a larger percentage. Correct the input meaning, then rerun the approved calculation.

Four changes should produce four explainable results

1. Change only the effective date. The current-year impact changes; the annualised local raise remains the same.
2. Change only the reporting rate. The translated forecast changes; the employee’s local-currency pay does not.
3. Change the pay basis. A monthly-to-annual mismatch is identified before approval. An hourly case exposes the working-hours assumption.
4. Change an approved input. Demonstrate which approvals must be revisited, who sees the change and what reaches the payroll export.

Require the final export to preserve employee identity, local amount, pay basis, currency and effective date. A dashboard screenshot without those fields is not enough evidence to hand the decision to payroll.

Require separate pay-rise and currency calculations

Keep a controlled spreadsheet if it can support your volume, access requirements and reconciliation. Buy compensation software when workflow, permissions or repeated manual work justify it. Neither format fixes an undefined budget policy.

Give shortlisted providers this same example and ask them to separate cycle approvals from Finance’s updated forecast. Reject the configuration if nobody can explain why the totals changed. Use the compensation software shortlist to choose trial candidates, then test their arithmetic and approval behaviour.

Questions buyers ask