Performance Based Pricing
Definition
Performance Based Pricing
Performance based pricing places an agreed share of the supplier’s fee at risk against measured performance, so the invoice moves up or down with results. The at-risk percentage is the whole mechanism, and everything else in the scheme is measurement design.
It differs from paying for an outcome — here the buyer still pays for the service and adjusts the price afterwards, rather than paying only when a result appears.
Most contracts set the band somewhere between five and twenty percent of fee. Below five it changes nothing; above twenty, suppliers price the volatility straight back in.
The design question is symmetry — a scheme with downside only is a penalty regime, and suppliers manage penalty regimes defensively rather than ambitiously.
Key takeaways
- A defined share of fee moves with measured performance, up, down or both.
- Symmetric schemes change behaviour; penalty-only schemes mostly change reporting.
- Measures must be inside the supplier’s control or the band is simply a discount.
- Baselines need re-setting periodically, or the incentive expires on its own.
How it works
Four decisions make the mechanism: which measures count, how much fee is at risk, whether the movement runs both ways, and how the baseline is reset. Getting the fourth wrong is why most schemes die in year three.
Public contracting has used formula-based adjustment for decades.
Federal rules state that where formula-type incentives apply, “increases in profit or fee are provided only for achievement that surpasses the targets, and decreases are provided for to the extent that such targets are not met”.
That is symmetry written as a rule. It is also the part commercial contracts most often drop, keeping the decrease and quietly removing the increase.
Measure selection is the second failure point. A supplier cannot control customer satisfaction driven by a product defect, and putting fee at risk against it buys resentment rather than effort.
| Design choice | Weak version | Version that works |
|---|---|---|
| At-risk share | Under 5% of fee | 10% to 15%, both directions |
| Measures | Six or more KPIs | Three, weighted |
| Control | Includes outside factors | Only supplier-controllable items |
| Baseline | Fixed for the term | Reset annually against evidence |
| Payment | Netted off silently | Reported and discussed monthly |
Baselines deserve particular attention. UK government guidance asks that contracts be designed “to incentivise delivery of the things that matter”, which stops being true once a target has been beaten for six consecutive quarters.
Three measures is usually the practical ceiling. Beyond that, weighting becomes arbitrary and the supplier optimises whichever measure is easiest to move.
Examples
Performance pricing changes behaviour where the measures are few, controllable and symmetric — and changes nothing but the paperwork everywhere else. The four cases below show both results, and one scheme still working in year four.
A utility puts 12% of fee at risk across three measures with upside and downside. First contact resolution improves by nine points in a year, and the supplier earns the upside twice.
A bank puts 20% at risk against eight KPIs with downside only. Reporting quality improves dramatically; operational performance does not move at all.
A retailer ties fee to customer satisfaction driven largely by delivery delays the supplier does not control. The band becomes a standing discount and the relationship sours.
A healthcare payer resets baselines annually against an agreed benchmark. The scheme is still producing improvement in year four, which is unusual and entirely deliberate.
Related terms
Several mechanisms link money to performance, and they differ in what triggers the payment. The entries below separate them, since the trigger is what determines supplier behaviour.
- Key performance indicator (KPI): the measure an at-risk band is written against.
- Service level agreement (SLA): the document holding the targets and their credits.
- Outcome based pricing: pays when a business result occurs, rather than adjusting a fee.
- Gain sharing outsourcing: splits a measured cost saving instead of adjusting a fee.
- Benchmarking: the evidence used to reset baselines credibly.
- Bonus or incentive compensation: the individual-level analogue inside the supplier.
- Vendor management outsourcing: the function that has to administer the scheme monthly.
FAQ
How much fee should be at risk?
Ten to fifteen percent is the range that changes behaviour without distorting pricing. Below five it is noise, and above twenty the supplier prices the volatility back into the base.
Should the scheme have upside?
Yes, wherever possible. Downside-only schemes produce careful reporting rather than better performance, because there is nothing to win by trying harder.
How many measures should there be?
Three, weighted. More than that and the supplier concentrates on whichever measure moves most easily, which is rarely the one that matters.
What if the supplier cannot control a measure?
Remove it. A measure outside supplier control converts the at-risk band into an arbitrary discount and destroys any incentive effect.
How often should baselines reset?
Annually, against evidence. A target beaten comfortably for a year has stopped being an incentive and become a guaranteed payment.
Is this the same as SLA-linked pricing?
Not quite. SLA-linked pricing runs credits against service level failures; performance based pricing can reward improvement above target as well.
Shortlist providers willing to accept symmetric at-risk terms through the Outsource Accelerator directory.
Independent
