What makes a good social value KPI?

A good social value KPI should leave very little room for interpretation.

It needs to be clear about what is being measured, who counts, where the activity should take place, what the target is, when it needs to be achieved, what evidence will be accepted and who is responsible for reporting and checking it.

That sounds straightforward, but in practice many problems with social value reporting start because one of those things was never properly defined.

Take a common commitment: create five jobs.

On paper, that looks perfectly clear. Once delivery starts, the questions begin.

Does a temporary role count? What about a role that already existed but has been refilled? Does the person need to live locally? How long do they need to stay in the job? And at what point is the job actually counted: when it is advertised, when someone accepts it, when they start, or after they have been in post for a certain period?

There is no single right answer to all of those questions. The problem comes when the buyer and supplier have different answers because nobody agreed the detail at the start.

This guide looks at how to avoid that.

What is a social value KPI?

A social value KPI is a way of monitoring whether a social value commitment made during procurement is actually being delivered.

It forms part of contract management. It gives both the supplier and the contracting organisation a shared measure of progress against what was promised.

Commitment, KPI, output and outcome

These terms are often used interchangeably, which can make reporting unnecessarily confusing.

A simple example helps.

The wider ambition might be to improve local employment.

The commitment made in the tender could be to create five additional jobs through the contract.

The KPI could then be the number of defined additional roles filled by eligible local residents by month 12.

The output is the five jobs once they have been filled.

The outcome is what happened to the people who got those jobs. Did they stay in employment? Did their income improve? Did they gain skills or greater financial security?

So, in simple terms:

A commitment is the promise.

A KPI is how you monitor that promise.

An output is what was delivered.

An outcome is the change that happened as a result.

A KPI can measure either an output or, where appropriate, an outcome. What matters is being clear about which one you are measuring.

What PPN 026 changes

PPN 026, published by the Cabinet Office on 5 August 2026, replaces the PPN 002 Social Value Model for central government procurements commencing on or after 1 January 2027.

The new Model focuses on two outcomes, Good Jobs and Skills, with minimum social value weightings of 10% for contracts between £1 million and £5 million and 20% for contracts worth £5 million or more.

The contract-management requirements are particularly important for KPI design.

In-scope organisations are expected to make sure that social value commitments made during procurement are monitored through appropriate contractual mechanisms and communicated to contract managers.

For contracts worth £5 million or more, at least one social value KPI should be set, alongside the wider KPI requirements under section 52(1) of the Procurement Act 2023.

Those KPIs are published on the central digital platform and reported through contract performance notices at least once every 12 months. Poor performance against social value KPIs may also become relevant when considering whether grounds exist to exclude a supplier from future tenders.

That makes the wording of the KPI more important.

A poorly designed KPI is no longer just an internal reporting nuisance. On a large central government contract, it can become a published measure of performance with wider consequences.

Local government and other parts of the public sector are not required to adopt the Model, although many already use similar approaches to social value commitments and contract KPIs.

Seven things a useful social value KPI should define

1. What exactly are you measuring?

Be specific about the thing you are counting.

A job start, a retained job, an apprenticeship start, an apprenticeship completion, a training completion, volunteering hours, local spend and an improvement in wellbeing are all different measures.

“Jobs and skills activity” is not really a KPI. It is a category.

The KPI needs to describe the specific event or change that will count towards the target.

2. Who counts?

If the commitment is intended to benefit a particular group, say who that group is.

Under PPN 026, this may include cohorts identified within the Model, such as young people not in education, employment or training, care leavers, and people with long-term health conditions or disabilities.

If there is no specific target group, that should be clear too.

Suppliers should not discover halfway through delivery that the buyer expected the activity to benefit a group that was never mentioned in the KPI.

3. Where does it count?

“Local” can mean different things.

It might refer to where the beneficiary lives, where they work, or the area in which the contract is being delivered.

Those definitions can produce very different results.

For example, a supplier might recruit people into a regional office located within the contract area, while the people filling those jobs commute from somewhere else entirely.

If geography matters, define it. In many cases that will mean specifying an area and checking an individual's postcode at the relevant point in time.

4. What is the target?

Targets can be expressed in different ways depending on the nature of the commitment.

A fixed number works well where the scale of delivery is predictable.

A percentage may be more appropriate where the target should rise or fall with the size of the contract. For example, you might measure the proportion of contract labour hours delivered by local residents.

A rate can work well for ongoing activity, such as apprenticeship starts per £1 million of contract value each year.

A milestone is useful for a one-off deliverable, such as launching a community programme by a specific date.

The important question is whether the target will still make sense if the scope or scale of the contract changes.

5. By when?

An end-of-contract target tells you very little until it may be too late to do anything about underperformance.

For longer contracts, interim milestones are usually more useful.

They allow both sides to see whether delivery is on track and make changes while there is still time to recover.

Those milestones do not necessarily need to be evenly distributed.

On a five-year contract, for example, employment targets might be higher during the years when workforce demand is expected to peak.

6. What evidence will you accept?

Evidence requirements should be agreed before delivery starts.

They should also be realistic.

Where possible, ask suppliers to provide evidence from records they already keep rather than creating a completely separate administrative process just for social value reporting.

If evidence standards are only discussed when the first report arrives, problems are almost inevitable. The supplier will provide what they have. The contract manager will then decide whether it is good enough, despite there never having been an agreed standard in the first place.

The table later in this guide gives some examples of suitable evidence for common social value KPIs.

For more on making evidence requirements proportionate, see What is proportionality in social value measurement?.

7. Who owns the KPI?

Someone needs to be responsible for delivering it and someone needs to be responsible for checking it.

Ideally, three roles should be clear:

• a supplier owner responsible for delivery and reporting
• a contract-management owner responsible for reviewing evidence and confirming performance
• an escalation route where performance falls behind or the two sides disagree

Without clear ownership, social value KPIs can easily sit in a contract for months or even years without meaningful review.

Worked example: turning “five jobs” into a useful KPI

The easiest way to see the difference is to take the same commitment and gradually make it more precise.

Version 1: Five jobs

This is too vague to monitor properly.

You cannot tell from the wording what counts as a job or when it should be counted.

A supplier could even argue that five advertised vacancies met the commitment.

Version 2: Five FTE jobs created through the contract

This is better.

Using FTE avoids counting part-time roles as whole jobs.

“Created through the contract” also introduces the idea of attribution. The supplier now needs to show that the roles are new and connected to delivery of the contract.

Evidence might include role-level records showing start dates, contracted hours and the contract to which each role is assigned.

Version 3: Five additional FTE roles attributable to the contract, filled by residents of the defined geography by month 12

Now the measure is much clearer.

“Additional” excludes jobs that already existed before the contract and were simply refilled.

The geographical requirement means residence needs to be checked.

The deadline creates a clear reporting point.

Evidence could include a pre-award workforce baseline, confirmation of postcode at the employee's start date, and a month 12 report.

Version 4: Add retention and job-quality criteria

You could take the measure further by specifying that the jobs must be paid at or above the Real Living Wage, offered on contracts lasting at least 12 months, and that at least four of the five people should still be employed at month 18.

At this point, the KPI is starting to move beyond simple delivery and towards the quality and sustainability of the outcome.

The evidence requirement naturally becomes greater too. You may need information on pay, contract type and continued employment at a later date.

That may be entirely reasonable on a £20 million contract where employment formed a major part of the social value offer.

For a much smaller contract, the additional reporting burden may not be proportionate.

Our guide to evidencing jobs and skills outcomes looks at this in more detail.

Should you measure an output or an outcome?

When an output KPI is useful

There is nothing wrong with measuring outputs.

If a supplier committed to starting 20 apprenticeships, the number of apprenticeships started is a perfectly legitimate contract KPI.

It is clear, countable and can be evidenced through enrolment records.

It tells the contract manager whether the supplier delivered what it promised.

When outcomes tell you more

You may also want to know what happened next.

How many of those apprentices completed their programme?

How many moved into sustained employment?

Did they gain qualifications or progress into better-paid work?

Those measures tell you more about whether the activity made a difference to the people involved.

They also take more effort to collect because they usually require follow-up after the original activity has taken place.

That may be worthwhile on larger contracts or where the commitment played an important role in the award decision.

The key is not to confuse the two.

“Twenty apprenticeships started” tells you that twenty apprenticeships started.

It does not, by itself, show that twenty people gained sustained skilled employment.

Problems arise when reports imply an outcome that was never actually measured.

How do you set a credible target?

A target should be ambitious enough to be meaningful, but realistic enough that it can genuinely be delivered.

Too low, and the social value weighting in the procurement is doing very little.

Too high, and the supplier either misses it or finds ways to deliver activity that looks good on paper but adds very little beyond what would have happened anyway.

A sensible target should take several things into account.

Contract value and duration affect what can reasonably be afforded and how much time there is to deliver.

The workforce requirements of the contract determine how many jobs or training opportunities may actually exist.

Previous delivery on comparable contracts can provide a useful benchmark.

The supplier market matters too. A target that only one bidder could possibly achieve may distort competition.

You also need a baseline. What does the supplier already do? Without that, it is difficult to judge whether activity is genuinely additional.

And then there is place.

Labour-market conditions, demographics, deprivation, skills levels and the availability of the people you are trying to reach all affect what is achievable.

They should also influence which social value outcomes are prioritised in the first place.

A commitment should respond to a genuine local need rather than simply choosing the activity that is easiest to count.

Place Explorer provides local need profiles that can help with this.

Our guide to setting targets using comparable and local evidence looks in more detail at how benchmarking and place-based evidence can be used together.

What evidence should sit behind a KPI?

The evidence should prove the specific thing the KPI claims happened.

It should also be proportionate to the scale and importance of the commitment.

One of the most common mistakes is using evidence from an earlier stage of a process to claim that a later stage has been achieved.

Where possible, keep an underlying record for each individual event, role, apprentice or payment, even if the final report only shows aggregate totals.

A total without any supporting record is very difficult to verify.

How do you prevent double counting?

Double counting becomes particularly important when suppliers are working across several contracts, programmes and corporate initiatives at the same time.

Contract activity versus corporate activity

A supplier may already run an apprenticeship programme, graduate scheme or long-standing charity partnership.

That can be useful evidence of the organisation's capability during a procurement.

It should not automatically be counted against a new contract KPI.

A useful question is: would this activity have happened without this contract?

If the answer is yes, it is probably part of the supplier's wider corporate activity rather than additional delivery created by the contract.

Delivery across several contracts

A supplier with several public-sector contracts in the same area may employ people, run apprenticeships or support community projects that could potentially be allocated to more than one contract.

Each activity should have a clear home.

The same five employees should not appear in the social value reports for three different contracts.

Supply-chain contributions

Activity delivered by subcontractors can legitimately count towards the prime contractor's KPI where the contract allows it.

But the evidence still needs to flow through.

There is also a risk that the same subcontractor activity is being claimed elsewhere, either by the subcontractor itself or against another client's contract.

If supply-chain activity is allowed, say so in the KPI and apply the same evidence standards to it.

Additionality and attribution provide useful sense-checks here.

You do not need to carry out a full SROI analysis for every contract KPI. But you should be able to ask whether the activity would have happened anyway and whether someone else is also claiming it.

What happens when a KPI is missed?

A missed target does not always mean the same thing.

Start by checking whether the problem is with the delivery or with the evidence.

A supplier may have created the jobs but failed to collect the agreed postcode evidence. That is an evidence issue.

If the jobs were never created, that is a delivery issue.

The next step is to understand why performance is behind.

Labour-market conditions may have changed. The contract scope may have shifted. Mobilisation may have been delayed. Or the original target may simply have been unrealistic.

Different causes call for different responses.

Where appropriate, agree corrective action and a revised timetable.

Keep a clear record of the shortfall, the explanation and what was agreed. Under PPN 026, that information may later be relevant to wider decisions about supplier performance, so both sides have an interest in making sure it is accurate.

Contractual remedies also need to be used proportionately.

A formal rectification process may be appropriate for persistent failure to deliver. It is unlikely to be the right first response to one missed milestone where there is a credible explanation and a realistic recovery plan.

Does meeting a KPI prove social impact?

No.

A KPI can tell you whether five people started the jobs that were promised.

It cannot automatically tell you whether those jobs lasted, whether those people would have found similar work anyway, or what changed in their lives as a result.

Those questions require outcome evaluation.

That may involve following up with the people affected, understanding what would have happened without the intervention and making adjustments for factors such as deadweight, attribution and displacement.

That is where approaches such as Social Return on Investment come in.

Both types of measurement are useful.

Contract management needs to know whether the commitment was delivered.

Outcome evaluation helps you understand whether it worked and what should be done differently next time.

Problems arise when one is presented as though it proves the other.

Social value KPI checklist

Before writing a social value KPI into a contract, ask:

1. Does it describe a specific event or change that can actually be counted?
2. Is it clear who is eligible?
3. Is the geography defined, including how it will be checked?
4. Will the target still make sense if the contract scope changes?
5. Are there useful interim milestones rather than just a final target?
6. Has the evidence standard been agreed in advance?
7. Can the supplier produce that evidence without creating an unreasonable reporting burden?
8. Are the supplier owner, contract-management owner and escalation route clear?
9. Does the KPI distinguish additional contract activity from existing corporate activity?
10. Is it clear whether supply-chain contributions can be included?
11. Is the KPI measuring an output or an outcome, and is that reflected honestly in reporting?

Use a one-page Social Value KPI Design Sheet to record the outcome sought, commitment, KPI, target, who counts, geography, evidence, reporting frequency and ownership. Then use the KPI Evidence Planning Worksheet to map the source, collection method, quality check, storage location and review status for each KPI. Share the completed worksheet with your team using one of these options:

Keeping commitments, evidence and outcomes in one place

Managing social value commitments across several suppliers or funded organisations? SVE Connect supports post-award social value reporting across multiple organisations, helping commissioners and delivery partners keep commitments, evidence and reported outcomes in one traceable process.