17 November 2026
The Milburn review: what it says and why it matters to anyone working in education
Alan Milburn's final report on young people and work came out today. The ambition is big: cut the NEET rate from 13% to 5% by 2036, a Youth Opportunity Service in every part of England by 2029, a named adviser and a personal participation plan for every young person who needs one, and a route into an apprenticeship, college place, supported employment or a job within four months.

Alan Milburn's final report on young people and work came out today. The ambition is big: cut the NEET rate from 13% to 5% by 2036, a Youth Opportunity Service in every part of England by 2029, a named adviser and a personal participation plan for every young person who needs one, and a route into an apprenticeship, college place, supported employment or a job within four months.
I've spent 25 years building digital products for education: councils, colleges, training providers, learning platforms. More recently I've been building one of my own, so I read the report partly as someone who's watched a lot of these schemes come and go, and partly as someone trying to build the tools that might make this one stick. Here's what stood out, and where I think technology can help, with Seren as one example of how.
The diagnosis is a data problem wearing a policy coat
The interim report in May said it plainly. Schools are judged on attainment, not destinations. Colleges are funded on enrolment and completion, not on whether someone stayed in work. Jobcentres see claimants. The NHS records treatment. Everyone holds a piece and nobody holds the whole.
The DfE already has a dataset (LEO) that links education records to tax, benefits and employment. Milburn called it one of the most powerful tools of its kind in the world. It isn't available to the schools, colleges or councils who'd actually use it. And 314,000 18 to 24 year olds in 2023/24 were in no system at all. Not working, not learning, not claiming. Nobody's job to find them.
So when the final report says every young person gets a named adviser and a participation plan, the first question I have isn't about funding. It's: what does that adviser see on screen? Where does the plan live? What's it made of?
A participation plan needs a common language
A plan that says "do a Level 2 in business admin" is a qualification plan. That's what we've had for 25 years. Milburn is explicit that the system prioritises qualifications over destinations and that destinations should be the primary aim.
A plan built for destinations has to start from a different place. What can this person actually do? What does the job or course they're aiming at actually need? What's the gap, and what's the shortest route across it?
You can't answer that without a shared vocabulary for skills. The UK has one now: the Standard Skills Classification. It sits alongside ESCO and O*NET internationally. Almost nobody's learning systems use any of them.
Seren is built on that vocabulary. Every learner, every piece of content and every goal (a job role, a qualification, something custom) is mapped to the same competency framework. That gives you the things an adviser needs and a conventional LMS can't do:
- Gap analysis between a learner and a goal, ranked by what matters most
- Reverse matching: start from what someone's good at and show them goals they're already close to
- Adjacency between goals, so a young person who's a few skills short of one route can see the others nearby
- An evidence loop, so completing something writes back to the learner's profile instead of just a certificate
- Cohort views, so a college or a mayoral authority can see where the gaps are across a group, not just one at a time
That's the plan Milburn's talking about, but as a live object rather than a Word document in a case file.
Four months is not long enough to build new courses
The four-month target is the detail I keep coming back to. The difficulty isn't time, it's the shape of the demand. Four months is per young person, and every one of them has a different gap between what they can do and what their route needs. You can't pre-write a course for every combination. What a provider can do is make the content it already has findable and matchable, so the right pieces can be pulled together for each plan when it's needed.
Colleges and training providers already own a huge amount of good learning content. The problem is it's stranded. SCORM packages from 2016. Moodle courses nobody can get into without an enrolment. H5P activities that only play inside one LMS.
Seren does two things with that. It can surface existing SCORM, xAPI and Moodle content in place via LTI, so nothing moves and nothing gets rebuilt. Or it lifts content across and rebuilds it as modern components, which we've already done for a client. Either way the content gets mapped to the skills framework, so it becomes something a plan can point at.
And because Seren is headless, that content can show up wherever the young person already is. In the college VLE. Inside Teams. On a council's website. Inside a partner's product. The review says the Youth Opportunity Service should bring together employment, careers, health and practical support. That won't be one system. It'll be six. A learning layer that sits across all six is more useful than a seventh login.
Measure retention, not starts
NFER's response to the report made the point I'd most like to see stick: judge this on retention and progression, not on entry. We've had Kickstart, New Deal, the Youth Guarantee. They counted starts.
The interim report had a stat on this: previously-NEET apprentices are less likely to still be with their employer afterwards, 49% against 68% for everyone else. Getting someone through the door is the easy half.
That's where the evidence loop matters. If every completed activity, every placement, every bit of feedback writes back to a skills profile, you can see progression as it happens. You can see who's stalling before they drop out. And a mayoral authority with a Participation Agreement to deliver on can see it across the whole area.
Where this is honest about its limits
Seren is a learning and skills layer. It won't fix a young person's mental health, pay their bus fare, or create an entry-level job in a town where retail and hospitality have shrunk. Milburn's report is right that those are the bigger forces.
What it can do is make the adviser's job possible. Give them a view of what someone can do, what they need, what's available, and whether it's working. Give the college a way to use content it already paid for. Give the strategic authority a picture across the whole cohort rather than a spreadsheet of starts.
The report isn't government policy yet. The response is due before Christmas, a White Paper early next year. But the shape is clear, and the 2029 deadline for the Youth Opportunity Service is not far off.
If you're a college, a strategic authority, a training provider or someone building the systems behind the Youth Opportunity Service and you want to talk about what a skills-anchored learning layer looks like in practice, let's chat.

