Get it right at the start: the Sourcing Playbook is a checklist for avoiding your next outsourcing failure
Post-mortems of failed outsourcing deals almost always find the same thing: the fatal decisions were taken before the contract was ever advertised. A market nobody assessed. A price nobody modelled. Risk dumped on a supplier who could never carry it. A first generation service outsourced with no pilot and no data. The Sourcing Playbook, now in its fourth iteration since 2019, is the government's accumulated answer to those post-mortems: 11 policies aimed almost entirely at the start of the process, because that is where value for money is actually decided.
Why the Playbook keeps saying the same things
Across four versions, the core has barely moved: pipelines, market assessment, validation reviews, delivery model assessments, Should Cost Models, pilots, KPIs, risk allocation, payment mechanisms, financial standing, resolution planning. That consistency is the point. These are not fashionable ideas, they are the recurring gaps the National Audit Office and industry keep finding when projects go wrong. Insufficient preparation time alone is one of the NAO's most reliable failure indicators.
Version 4, published in June 2023, refined rather than reinvented: better TUPE content, guidance on applying policies through frameworks and call-offs, more help for local government, and a serious treatment of inflation and indexation, the issue consultations said practitioners needed most. A playbook that changes its fundamentals every two years would be admitting they were wrong. This one is doubling down.
The policies everyone skips
In practice, three policies get skipped most often. Pipelines are published late, thin or not at all, which quietly excludes the SMEs and voluntary sector organisations that cannot mobilise a bid team in three weeks. Delivery model assessments are written to justify a decision already taken, rather than to make one. And pilots are dropped for timetable reasons on exactly the projects that need them: first generation outsourcing, where nobody, including the buyer, yet understands the requirement, the data or the risks.
The Playbook's counter is the shall versus should distinction. Complex outsourcing, first generation services, major transformations, thin markets, novel or contentious requirements, triggers hard obligations: a Project Validation Review, a Should Cost Model, a pilot, embedded Cabinet Office support. If your project meets the definition and you are not doing these things, you are not exercising judgement, you are accumulating explanations for a comply or explain conversation you will eventually have to have.
It is not a commercial document, and that is deliberate
One of the Playbook's quietest but most important arguments is that sourcing has failed when treated as a purely commercial enterprise. Its role matrix expects finance to co-own Should Cost Models, project delivery to co-own pilots and KPIs, and policy teams to be in the room from the start. Several key policies are explicitly expected to be led by non-commercial professionals.
For buying organisations, the practical implication is structural: if your sourcing decisions are made by a commercial team alone and reviewed by nobody, you are organised for the failures the Playbook was written to prevent. The fix costs meetings, not money: joined-up teams at policy stage, functional experts at the validation review, and decision makers who see the delivery model assessment before the answer is presumed.
Wider public sector: adopt it proportionately, but adopt it
The Playbook is mandatory for central government, but version 4 went out of its way to invite local government and the wider public sector in. The invitation is worth accepting, with judgement. A district council re-letting a grounds maintenance contract does not need the Complex Transactions Team, but it absolutely benefits from a market assessment, an honest make versus buy discussion, a whole life cost estimate and KPIs that measure what residents actually experience.
The mechanics are the same at any scale: know your market, know your costs, allocate risk to whoever can manage it, and leave an evidence trail that shows the decision was made properly. Buyers who work that way get better bids from better suppliers, because the market can tell the difference between an organisation that has read its own playbook and one that is improvising.
The takeaways
- Value for money is decided in preparation and planning, which is why 5 of the 11 policies bite before publication.
- Complex outsourcing triggers hard requirements: PVR, Should Cost Model, pilot and embedded support.
- Pipelines, delivery model assessments and pilots are the most skipped policies, and the most costly to skip.
- Sourcing is cross-functional by design: commercial teams alone cannot deliver the Playbook.
- Wider public sector buyers should adopt the policies proportionately rather than ignoring them.
Want the full breakdown?
The complete explainer covers the key facts, the requirements in detail and a practical action list, free and printable in the Procurement Library.