Service contracts and SLAs: what the agreement should actually say

Every imaging system, ventilator and laboratory analyzer eventually needs service, and the contract behind that service decides how quickly a department recovers from failure. Two models dominate: a full-service agreement that bundles parts, labor and preventive maintenance into a fixed fee, and pay-per-call work billed after each visit. Neither is automatically cheaper. The right choice depends on device criticality, on how the equipment fails and on how much uncertainty the organization can absorb in a budget year.

Two ways to buy service

A full-service contract transfers risk to the provider. The organization pays a predictable amount and stops worrying about the cost of a failed high-voltage generator or a replacement detector. The provider, in turn, prices that risk into the fee, which means the contract costs more in a year when nothing breaks.

Pay-per-call keeps money in the budget during quiet periods and exposes it when equipment fails. It suits stable, low-acuity devices with cheap parts, and it punishes organizations that own a small number of highly specialized systems where a single failure can consume an entire service budget.

What a full-service contract actually covers

The label full service means very little on its own. One agreement includes parts, labor, travel, software updates and unlimited corrective visits. Another includes labor only, charges for consumables and excludes anything the vendor classifies as customer damage. Both may be sold as comprehensive cover.

Read the exclusions before comparing prices. Detectors, probes, transducers, batteries, uninterruptible power supplies and user-induced damage are the usual gaps. So is the difference between a software patch and a software upgrade, which matters more as devices become connected and their value shifts into firmware.

Where pay-per-call makes sense

Time-and-materials work fits devices with a long history of stability, a healthy supply of spare parts and in-house staff who can handle first-line checks. It also fits equipment nearing replacement, where a multi-year agreement would outlive the device and lock in spending on hardware that is about to be retired.

For planning, treat pay-per-call as a variable cost and model it from the last few years of actual invoices rather than from the vendor's estimate. That history is also useful when negotiating a full-service price, because it shows the provider what the account has genuinely cost.

The clauses that decide the outcome

Price and coverage attract attention, but the operational clauses decide whether the contract works at two in the morning. These deserve the same scrutiny as the fee schedule, and they are usually where a service organization wins or loses the argument later.

• Definitions of response, arrival and repair, stated separately and in hours from the logged call.
• Uptime commitment measured on clinical availability, not on vendor business hours.
• Parts availability, including a statement on obsolete components and last-time buys.
• Who may perform work, and whether in-house staff can do first-line tasks without voiding cover.
• Software and cybersecurity support, including who applies updates and who confirms the device is safe to reconnect.
• Escalation path with named roles, not a generic support queue.
• Termination terms, notice periods and what happens to prepaid visits at exit.

Response time is not repair time

Contracts often promise a fast response and say nothing useful about restoration. A provider can respond in minutes by acknowledging a ticket and still leave a device down for days waiting for a part. The distinction matters most for equipment on a critical care pathway.

A sound agreement defines each measure, names the system of record and states how disputes are resolved when the vendor's ticket log and the department's experience disagree. Without that, performance reviews turn into arguments about definitions rather than about equipment.

How to verify performance

Verification needs data the organization controls, not only the provider's monthly report. Work orders should be raised in the CMMS at the moment of the call, with timestamps for the request, the arrival and the return to service. That record is what makes a service review factual.

Preventive maintenance visits need the same treatment. Confirm that the tasks on the vendor's checklist match the manufacturer's schedule, that measurements are recorded rather than ticked, and that the device is returned with a functional safety check. A signed visit sheet with no values on it proves that someone attended, not that anything was tested.

Review performance quarterly against the indicators already used for the department, so contract reporting and internal HTM KPIs and benchmarking tell the same story. Consistent measures also make the renewal conversation about evidence instead of impressions.

Renewal and exit

Renewal is the moment of maximum leverage, and it arrives faster than most departments expect. Start the review while there is still a full quarter of performance data and enough time to test the market, and bring the equipment's remaining life into the discussion. A contract priced for a device that will be replaced within the term is money spent on the wrong asset.

Feed the outcome into equipment lifecycle budgeting rather than treating service as a fixed annual line. When cover, failure history and replacement timing are reviewed together, the contract stops being a purchase and becomes part of the decision about how long the device should stay in service.