Buy the Outcome, Not the Timesheet: Moving Workplace Service Contracts Off the Clock
Photo: NAVFAC, CC BY 2.0, via Wikimedia Commons
Procurement teams have spent years squeezing unit rates on cleaning, security, and front-of-house coverage. The result is often a cheaper hour, not a better outcome. A building can hit its contracted labor hours every week and still have conference rooms that fail inspection, lobby coverage that thins during the 8:45 a.m. arrival peak, and incident reports that arrive three days late. Paying for inputs rewards presence. It does not reward performance.
Why the Hours Model Persists
Time-and-materials pricing is easy to budget, easy to compare, and easy to defend in a bid review. Three vendors quote a rate per hour and the lowest number wins. The model also suits the vendor, who is paid whether the work is well sequenced or not. It suits the buyer's finance team too, because a fixed monthly invoice is simpler than a variable one tied to a scorecard.
The cost of that simplicity shows up in three places: overstaffing during predictable lulls, understaffing during predictable peaks, and a chronic inability to prove whether a service is actually working. When a regional manager asks why a building feels under-serviced, the answer is a staffing schedule, not a result.
What Outcome-Based Actually Means
Outcome-based contracting is a spectrum, not a binary. At one end sits performance-based pricing, where a portion of fees is placed at risk against defined service levels. At the other end sits true outcome pricing, where the vendor is paid against a business result the buyer cares about: cleanliness verified by inspection scores, security measured by response time and incident closure, concierge measured by visitor wait time and host satisfaction.
Most US enterprises are not ready for the far end of that spectrum, and they do not need to be. A practical starting point is a hybrid: a fixed base fee for core coverage, plus a variable pool, typically 10 to 20 percent of contract value, tied to a small set of well-defined KPIs.
Choosing KPIs That Survive Contact With Reality
The most common failure is a scorecard full of metrics no one can measure consistently. A KPI must be defined, observable, and attributable. Cleanliness is not a KPI. A pass rate on a defined inspection checklist, sampled weekly by a third party, is.
Useful examples by service line:
- Janitorial: inspection pass rate, restroom restock compliance, response time to spill calls, floor-care cycle adherence.
- Security: average response time to alarms, patrol checkpoint completion, incident report turnaround, access-control exception resolution.
- Concierge and front desk: visitor check-in time, host notification latency, visitor satisfaction score, meeting-room reset completion.
- Maintenance: work-order first-time fix rate, mean time to repair, preventive maintenance completion.
Keep the scorecard to five to seven measures. A vendor cannot manage twenty, and a buyer cannot govern twenty.
Contract Language That Holds Up
The clauses matter more than the concept. At minimum, the agreement should specify:
- KPI definitions and measurement methodology, including sampling approach, audit frequency, and who pays for measurement.
- A baseline period, typically 60 to 90 days, during which performance is measured but not penalized.
- A remedy ladder: self-corrective action plan, service credits, and termination for persistent failure.
- A gain-share mechanism for savings or performance improvements the vendor helps generate.
- Clear exclusions: events outside the vendor's control, changes in building occupancy, and buyer-caused delays.
- Dispute resolution with a defined evidence standard.
A workable service-credit formula caps exposure. For example: each KPI miss below threshold triggers a credit of 1 percent of the monthly variable pool, capped at 15 percent, with two consecutive quarters of failure creating a termination right.
A Scorecard You Can Actually Run
| KPI | Weight | Target | Measurement | Consequence |
|---|---|---|---|---|
| Inspection pass rate | 30% | 95% or better | Weekly third-party sample of 40 spaces | Credit below 90% |
| Response time, priority 1 | 25% | 15 minutes or less | CMMS timestamps | Credit below 80% compliance |
| Visitor check-in time | 15% | 90 seconds or less | Badge system median | Credit below target |
| Meeting-room reset | 15% | 98% or better | Random daily audit | Credit below 95% |
| Satisfaction score | 15% | 4.3 out of 5 | Quarterly occupant survey | Credit below 4.0 |
Pitfalls to Avoid
Baseline gaming. Vendors may perform well during the baseline period and ease off afterward. Use a rolling baseline or benchmark against peer buildings.
Underpriced risk. A vendor who accepts aggressive KPIs will price the risk into the base fee. Compare total cost of ownership, not just the variable pool.
Measurement disputes. If the buyer and vendor cannot agree on how a KPI is measured, the KPI is decorative. Write the method into the contract and test it in the first quarter.
Scope creep without re-pricing. Outcome contracts still need a change-order process. New floors, new services, and occupancy swings should trigger a documented adjustment.
One-sided risk transfer. A contract that pushes all risk to a vendor who cannot control the outcome, such as cleanliness in a building with a failing HVAC system, will produce disputes rather than performance.
A Phased Path
Start with one service line, usually janitorial or security, where outputs are observable and measurement is cheap. Run a six-month pilot with a modest variable pool, publish the scorecard monthly, and use the results to refine definitions before scaling to hospitality and maintenance. The goal is not a clever contract. It is a service model where the buyer can answer a simple question, did we get what we paid for, with evidence rather than an invoice.