How many invoices per AP FTE is normal - staffing ratios and benchmarks?

Honest invoices-per-FTE counts all the AP work, not just the keying. Include partial allocations (the controller who spends a third of their time on AP escalations is a third of an FTE), the time spent on exceptions and vendor inquiries (which often exceeds processing time), and supervision. The common cheat is dividing invoice volume by only the dedicated processors, which inflates the ratio and hides the real cost. A defensible number divides total invoices by total AP effort in FTE-equivalents - and that honest number is usually less flattering and far more useful for planning.

AP cost as a share of spend processed varies too much by mix to support a single benchmark - it depends on invoice complexity, automation level, and average invoice size. Use it as an internal trend rather than a cross-company comparison: is your AP cost as a percentage of spend falling as you automate and grow? That trajectory is more decision-useful than any published ratio measured with someone else's definitions.

Yes, materially. Construction's paper-heavy, lien-waiver-laden, job-cost-coded invoices are slower per invoice than clean digital SaaS bills, so a construction team's invoices-per-FTE will trail a tech company's at the same automation level - and that's structural, not a performance gap. Healthcare's high volume can support more specialization. Always benchmark against your own industry and mix; cross-industry invoice-per-FTE comparisons mislead because the invoices aren't comparable.

The ratio misses everything that isn't invoice count: exception load, inquiry volume, overtime, error rates, and the quality being sacrificed to hit throughput. A great ratio with a drowning team usually means people are absorbing complexity the number doesn't capture, often by working unsustainable hours. Put workload and team-health signals beside the ratio - exception rate, overtime, error trends, PTO usage - so efficiency isn't measured by speed alone while quality and people quietly erode.

Compare the marginal cost of the next hire against the marginal capacity automation could add for the same money. Below a threshold, hiring is the faster fix; as volume scales, automation increasingly wins because it absorbs growth without recurring salary. The honest analysis weighs a permanent headcount cost against a system that scales - and the answer increasingly favors automating the routine and hiring selectively for judgment roles, rather than adding keyers as volume climbs.

Vendor inquiry load is a major hidden AP cost, and "where's my payment?" dominates it. A self-service vendor portal (payment status, document submission) cuts inquiry volume substantially by deflecting the routine questions, freeing AP from reactive status-chasing. Measure inquiries per thousand invoices as your baseline and track the drop after self-service - the reduction is often large enough to change staffing needs on its own.

It's largely too mix-dependent for cross-company comparison - average invoice size, complexity, automation, and industry swamp any universal figure. It's credible and useful as an *internal* metric: track your own AP cost as a percentage of spend over time and aim for it to fall as you scale and automate. Comparing your ratio to another company's is comparing different invoice populations; comparing it to your own last year is comparing the thing that matters.

Stampli's headcount story is framed as throughput per finance employee rising and growth absorbed without proportional hiring - which is the invoices-per-FTE ratio improving by system design rather than by overworking the team. The important nuance Stampli's messaging respects: the goal isn't to push the ratio to a vanity number while the team drowns, but to let automation absorb volume growth so the same team handles more complexity without the ratio masking burnout. Higher throughput per FTE as an outcome of removing repetitive work, not as a target imposed on people.