Benchmark Clause
Definition
Benchmark Clause
A benchmark clause is a contract provision that lets one party test the deal’s prices or service levels against a comparable market sample during the term. It creates a review right, not an automatic price cut, and the difference between those two decides everything.
Long outsourcing contracts date quickly. A rate agreed in year one can sit well above the market by year four, and the buyer has no orderly route back to a fair price unless a mechanism was written in advance.
The clause supplies that route. It names who may call a benchmark, how often, who performs it, what the comparator set must look like, and what happens if the exercise shows the price is out of line.
Buyers rarely use the right as often as they negotiate for it. Calling a benchmark signals distrust, so the mechanism sits unused in most contracts until a renewal is already in view.
Most benchmark clauses fail on the last point. They produce a finding and then require the parties to negotiate in good faith — a polite way of saying that nothing has to change.
Key takeaways
- A benchmark clause tests contract pricing against a comparable market sample mid-term.
- The comparator definition matters more than the trigger or the frequency.
- Clauses that end in “negotiate in good faith” rarely move the price.
- An adjustment formula tied to the finding is what makes the clause bite.
How it works
One party serves a benchmark notice. An independent benchmarker agrees a comparator set, normalises for scope and volume differences, reports a market range, and the contract then says what the parties must do with that number.
Normalisation is where disputes start. Two contracts are never identical in volume, service level, geography or transition history, so the adjustments made before comparison can move the answer more than the raw prices do.
Public procurement guidance treats the same problem as a pricing question — one of comparison rather than negotiation. The Federal Acquisition Regulation directs that price analysis demonstrate a price is reasonable.
The test is whether the price stands up “in comparison with current or recent prices for the same or similar items, adjusted to reflect changes in market conditions”. That adjustment step is the whole argument.
| Design choice | Weak drafting | Stronger drafting |
|---|---|---|
| Who may trigger | Either party, any time | Buyer, annually after year two |
| Comparator set | “Similar contracts” | Named size, geography, scope band |
| Benchmarker | Agreed later | Named firm or panel in the schedule |
| Outcome | Good-faith negotiation | Automatic adjustment to upper quartile |
| Cost | Silent | Split, or loser pays |
Examples
Benchmarking appears across service lines, and its usefulness tracks how commoditised the work is. The three cases below are typical of what buyers actually get when the clause is exercised.
A seat-based contact centre deal benchmarks cleanly because rates are widely published and scope is comparable. This is the territory where benchmarking produces a defensible number quickly.
Public templates assume this kind of review is routine. The Cabinet Office Model Services Contract is drafted for services contracts valued over £20 million, where mid-term price testing is expected rather than exceptional.
A finance and accounting arrangement benchmarks less cleanly, because process maturity varies enormously. Buyers here often pair the clause with an indexation clause so that inflation is handled separately from market drift.
A bespoke application-support deal barely benchmarks at all — and buyers should stop pretending otherwise. There is no comparable sample, so buyers rely instead on a value-based pricing outsourcing structure and on open-book cost visibility.
Related terms
Several provisions adjust price or performance after signature, and they are routinely confused with one another. The list below draws the line between each of them and a benchmark clause, because the comparison target is what separates them.
- Benchmarking clause outsourcing: the same mechanism described from the outsourcing-contract angle.
- Most favored customer clause: a promise about the supplier’s other customers, not about the market.
- Uplift clause: a permitted increase, which is the mirror image of a benchmark reduction.
- Service level agreement SLA: the performance side that a benchmark can also test.
FAQ
How often should a benchmark be allowed?
Once every twelve to eighteen months after an initial stabilisation period is common. More frequent rights create administrative load without producing new information.
Who pays for the benchmarking exercise?
Usually the party calling it, sometimes split. A loser-pays arrangement discourages speculative notices but also discourages legitimate ones.
What makes a comparator set defensible?
Named boundaries agreed in advance: contract size band, geography, service scope and vintage. Leaving the set to be agreed at the time guarantees an argument.
Can the supplier trigger a benchmark?
Yes, where the contract says so. Suppliers occasionally want the right when they believe their rates have fallen below the market and an increase is justified.
Does a benchmark clause apply to service levels too?
It can. Performance benchmarking compares the contracted targets against what comparable providers now commit to, which matters most in long deals.
What happens if the parties cannot agree on the result?
The dispute route in the contract applies, usually expert determination rather than litigation, because the question is technical and the parties want it settled fast.
Find providers who price transparently and accept mid-term review in the Outsource Accelerator directory.
Independent
