The Standing Agreement and How It Differs From a Service Contract
Why this matters
An institution will hand you two pieces of paper that look similar and are not. One is a standing agreement, sometimes called a blanket purchase order, a term contract, a master agreement or an on-call contract. The other is a service agreement covering scheduled maintenance. Shops routinely price the first the way they price the second, discover it generates a third of the hours they assumed, and then either walk away from a good account or hold it at a rate that loses money on every call. The two instruments sell different things. One sells availability, the other sells work. Get that straight and both become priceable.
What each one actually buys
| Standing agreement | Service agreement | |
|---|---|---|
| What the institution is buying | The right to call you and have you show up, at pre-agreed rates, without issuing a new procurement each time | A defined scope of work performed on a defined schedule |
| Volume obligation | None. A ceiling, not a floor | Yes. The scope is the volume |
| What triggers work | A release, task order or work order against the agreement | The calendar |
| What you can forecast | Nothing reliably, unless history gives you a range | Hours, months in advance |
| Typical term | One year with renewal options, often 3 to 5 total years | One year, auto-renewing |
| What you are pricing | Availability plus a rate schedule | Hours plus materials plus a margin on a known scope |
The single word that separates them is obligation. A standing agreement is an option the institution holds and you wrote. They may exercise it forty times or twice, and nothing in the document requires either.
The gate
Run every institutional agreement through one question with two parts, and answer both before you price anything:
Does this document obligate volume, and does it define the scope in advance?
- Both yes: it is a service agreement. Price it on hours, because you know the hours.
- Both no: it is a standing agreement. Price it on rate plus whatever response commitment you are making, because the hours are unknown and the commitment is the real product.
- Split: it is a hybrid, and the two halves get priced by their own rule rather than blended. These are common and are covered after the two cases.
Case A: a blanket purchase order at a school district
The district issues a one-year blanket purchase order with a not-to-exceed ceiling, a rate schedule for regular and after-hours labor, and a stated response commitment of 4 hours for anything affecting an occupied classroom.
Run the gate. Volume obligation: none. The ceiling caps what they may spend and promises nothing. Scope defined in advance: no; scope is whatever breaks. Both no, so this is a standing agreement.
Price the response commitment, because that is what you are actually selling. Work out what it costs you in capacity:
- The shop runs 5 technicians at 40 scheduled hours, so 200 technician-hours a week.
- District history, from the prior vendor's records or your own first quarter, shows draw ranging 12 to 30 hours a week with a mean near 18.
- To honor a 4-hour response on a 30-hour week without breaking scheduled work elsewhere, you have to keep about 30 hours a week either unsold or interruptible. That is 15% of the 200 available technician-hours.
- On an average week, you draw 18 of that reserved 30. Utilization on the reserved block is 60%, so about 12 hours a week, or 6% of total technician-hours, is capacity you are holding and not selling.
That 6% is the price of the option, and it does not appear anywhere on the rate schedule. Two ways to recover it, and you should choose deliberately rather than drift:
- Carry a premium on the standing rate over what you would charge the same customer for scheduled work, sized to the reserve you are actually holding.
- Make the reserve interruptible rather than idle by filling it with work you can push a day, which converts the cost from 6% of capacity to whatever the rescheduling actually costs you in efficiency and customer goodwill.
The second is the better answer for most shops and it only works if the interruptible work is genuinely interruptible. Filling the reserve with a homeowner install that cannot move is the same as having no reserve at all, and the failure mode shows on the first week you draw 30 hours: you miss the response commitment, and a missed response on an institutional agreement is the finding that appears in the renewal review.
Case B: an annual maintenance agreement at a hospital
The same shop is offered an annual preventive maintenance agreement covering a fixed equipment list, with quarterly visits and a scope defined per visit.
Run the gate. Volume obligation: yes, the scope is the volume. Scope defined in advance: yes, down to the equipment list. Both yes, so this is a service agreement and it prices on hours.
The estimating problem here is not forecasting. It is that the hours are real and knowable, and the risk is entirely in the scope wording. Two clauses decide whether this agreement is profitable:
- What "covered" means when the visit finds a fault. If the agreement covers inspection and minor adjustment but a failed component is billed separately, you need the boundary written in terms someone can apply on site, not the word "minor." A workable boundary is a stated labor threshold per finding plus an explicit list of consumables included.
- What happens when access is denied. In a hospital you will be turned away from a space because it is in use. If the agreement obliges four visits and does not say what happens to a visit you could not perform, you will perform five and bill four.
Both cases involve the same institution type, the same crew and the same rate card. They resolve oppositely because one sells an option and the other sells work.
Where the split answer shows up
Most real institutional agreements are hybrids, and the honest move is to price each half by its own rule rather than blending them into one number.
- Scheduled scope with an on-call rider. Price the scheduled scope on hours; price the rider on reserve. If the rider carries a response commitment, it costs capacity whether it is used or not.
- A ceiling with a stated minimum. Some institutions guarantee a floor to keep vendors interested. That floor is a real volume obligation and prices like scheduled work up to the floor, with everything above it priced as option.
- Multiple award, no exclusivity. Where the institution holds identical standing agreements with three shops and dispatches to whoever answers, you are holding reserve for work that may go to somebody else. That changes the reserve math badly and is worth pricing at a lower response commitment rather than the same one. See related: Being One of Several Vendors on the Same Building.
The terms that decide whether the agreement is worth holding
Read for these six before you sign anything. Each one is a number or a named condition, not a sentiment:
- The not-to-exceed ceiling and what happens at it. When the ceiling is reached, work stops until the ceiling is raised, and raising it is a procurement action with its own calendar. Ask what the ceiling was on the prior year and how often it was raised.
- Whether a release or task order is required per call. If yes, that document is your authorization, and work performed without it is exposed. See related: Why Work Done Without a Purchase Order Often Goes Unpaid.
- Response commitment and how it is measured. From what event, to what event. "Four hours to respond" measured from ticket creation is different from measured from your acknowledgment, and the two can differ by a whole overnight.
- Rate escalation over the term. A three-year agreement at year-one rates is a real cost you are absorbing if labor costs move, and most institutions will accept a stated annual escalator if it is proposed at bid rather than requested mid-term.
- Payment terms. Institutional terms in the net-30 to net-60 range are normal and they are a working-capital commitment, not a nuisance. See related: Net Terms and What They Do to a Small Shop's Cash Position.
- Term, renewal options and the notice period. Know the date the re-bid starts, because it is usually earlier than the expiration date. See related: The Annual Re-Bid and How to Be Ready for It.
The trap: staffing to the ceiling
The most expensive mistake with a standing agreement is treating the ceiling as a forecast. It is a spending authorization, sized by a buyer to cover a bad year, and on the average year it will be substantially underspent.
A shop that hires against the ceiling ends up carrying the payroll of a technician the agreement never funds. The correct read is the opposite: staff to the historical mean draw, hold the reserve with interruptible work, and treat any month above the mean as overtime or subcontracted rather than as the new baseline. Ask for the prior two years of actual spend under the agreement before you sign. A facilities department will usually provide it, and if they will not, the prior vendor's invoice history is often a public record at a public institution.
Where this flips: on a first-year agreement at an institution that has just brought a building online or has just deferred maintenance through a lean budget cycle, the historical mean is not a guide, because there is either no history or a backlog waiting to be released. In that specific case, price conservatively, watch the first two quarters, and renegotiate rather than staff.
How to verify you got this right
- Read your own agreement and answer the gate out loud. If you cannot say whether it obligates volume, that is the first question to send to the buyer, in writing, before you price.
- After two quarters, compare actual hours drawn against what you assumed and against the reserve you held. If drawn hours are running below half your reserve, your response commitment is over-bought and you should either sell the slack or trade the commitment for a better rate at renewal.
- Check that every call in the period had its release or work order number recorded on your invoice. A missing number is the most common reason an institutional invoice sits unpaid, and it is invisible until the aging report shows it.
- Confirm your ceiling burn rate quarterly. Reaching the ceiling in month eight with no path to raise it is a four-month gap in an account you built capacity around.
References
- See related: What a Purchase Order Actually Obliges Both Sides To, Why Work Done Without a Purchase Order Often Goes Unpaid
- See related: Net Terms and What They Do to a Small Shop's Cash Position, Being One of Several Vendors on the Same Building
- Federal Acquisition Regulation, Part 16, describing indefinite-delivery and task-order contract structures that state and local term contracts commonly mirror
- Trade-standard practice for on-call and term service agreements in facilities contracting