Phase 1 - Define

1.7 Model scenarios to price outcomes

Rate Cards

Where the SIB has been originated by the outcomes payer, they may publish a Rate Card: a list of outcomes for which they are willing to pay, with a value attached to each.

When commissioners publish Rate Cards, they can either be:

  • fixed – weighting the procurement process towards quality, or
  • flexible – in which case commissioners expect discounts to it and may select based on value for money.

The rates are typically calculated from models of long-term savings for the commissioner, and the cost of other comparable programmes. Sometimes commissioners allow for providers to share in any cost savings they achieve against the cap. 

In other cases, where you as a provider or intermediary are leading the SIB development, there are no set rates. Rather than responding to an invitation to bid, you will need to negotiate outcomes from scratch or you may be asked to provide your own Rate Card.

Pros and cons:

  • Rate Cards fix some variables, leading to a potentially simpler SIB design process that makes the question “Can we make these numbers add up?” easier to answer.
  • On the other hand, without a pre-defined Rate Card, you are free to design a SIB that works for you both organisationally and financially.
  • The negotiating of outcomes rates and definitions can also be a crucial step in a developing a strong provider-commissioner relationship.

Starting from scratch:

“Our process was way more interactive, a true partnership. It was brilliant but hard. It pushed out loads of questions about who has responsibility for what.”

– Provider who defined their outcomes without a Rate Card

It can be helpful to refer to historical Rate Cards, such as those from the central government SIB funds:

DWP Innovation Fund Round 1 rate card

Refining outcomes projections

Scenario modelling often depends on a sometimes hidden factor: the details of the outcome indicators and the mechanism for verifying that success. Many who have been through the process wish that, even at this early stage, they had tested their numbers by sketching the process of identifying and claiming outcomes to avoid surprises.

You will already have discussed cohort-level vs individual-level outcomes and agreed an outline definition. Modelling for volumes and rates can involve further complicated conversations about:

  • Whether to downscale expectations if success relies on establishing a relationship or data sharing agreement with a specific authority or data source by way of validation

  • Whether to downscale expectations if success relies on keeping in contact with a potentially chaotic beneficiary group

  • Whether there is likely to be a time lag in establishing success – for example, if annual results are published by a statutory body – which could affect cashflow

  • What volumes would constitute statistical significance for the cohort size being discussed

  • Where outcomes are not binary – where they call for an improvement or increase or decrease in something – how the scale of change required will affect success rates. For example, how many outcomes can you expect if a 10% improvement is required compared with a 50% improvement?

  • Whether you are engaged in a zero-sum game: will success in one area affect your ability to achieve outcomes in another?

  • Whether external political, economic, social or technological factors could interrupt your success rates

In summary, we all need to build human error, reluctance, confirmation bias and even deceit as well as healthy optimism into our models!

Pricing outcomes

When discussions with a commissioner become promising, before starting to negotiate rates for outcome payments, it is prudent to model different scenarios, especially a down-side case to protect you from underpricing your outcomes.


Scenario analysis model
Contributed by Sune Frandsen, KKR

Download Tool

1. Start with your assumptions from Section 1.4

  • Drivers: duration of programme, number of participants recruited, percentage of participants dropping out, percentage of participants achieving each outcome, case load per staff member

  • Costs: delivery costs, core costs, SIB-related costs

2. Link the costs to the drivers

  • E.g. staff costs depend on number of participants, case load per staff member

3. Link the outcome payments to the drivers

  • E.g. outcome payments depend on number of participants, percentage of participants dropping out, percentage achieving each outcome and the Rate Card

4. You should be able to change the drivers and see the effect on the costs and the outcome payments

5. Model four different scenarios, from “best case” to “downside case”. Input the appropriate numbers as drivers in each case and see the effect on costs and outcome payments.

  • Best case: what is the theoretical maximum value of the contract if everything goes well? Does the contract apply a cap?

  • Expected case: an ambitious but achievable view of the programme

  • Break-even case: what is the minimum level of outcomes needed before the contract makes a loss?

  • Down-side case: take a more cautious view of e.g. the number of people who complete the programme, or the staffing levels needed. e.g. programme is delayed, recruitment is slow, half of the beneficiaries drop out, staff fall ill, etc.

This analysis will help you to identify a reasonable outcome payment level that covers costs, even if things do not go exactly to plan. In addition, it will help you to identify the key risks of the contract and decide how to manage them, and it will show how much investment might be needed.

This will usually form one of a series of stage-gates for a ‘go’ or ‘no go’ decision, defined for itself by each potential delivery organisation.

Further resources

Next: Phase 1 Checklist