What the Sats Marking Crisis Tells Us About Third-Party Contract Handovers
Every July, Year 6 pupils across England find out how they did in their Sats. This year, a lot of schools found out how their exam board did instead, and the answer wasn’t good.
Pearson took over the £180 million Sats marking contract from Capita this year and rolled out a new marking platform, ModMark, at the same time. What followed was weeks of technical problems in the run-up to results day, originally scheduled for 7 July: markers reporting questions that took twenty minutes to load, allocations that doubled overnight, and an already-extended deadline that still wasn’t enough. Pearson ended up asking maths markers to help clear a backlog of grammar, punctuation and spelling questions just to get through the volume. Results have now been pushed back to 16 July, and one education union has called for the entire assessment series to be brought back into the public sector.
I want to be careful here, because it’s easy to pile onto a story like this with hindsight and certainty neither of which anyone actually had in the room when the decisions were made. But strip away the headlines and there are three separate failures sitting on top of each other, and each one is worth looking at on its own, because each one is completely avoidable and each one shows up in client work I do on a regular basis.
A Change Management Failure
Migrating a live, high-stakes, statutory national assessment process onto a brand new platform, in the same year you take over delivery of the contract, is about as high-risk a piece of change management as you could design if you were trying to create one deliberately. Two major variables changed at once: who’s running it, and what system they’re running it on. Best practice in change management is to isolate variables wherever you possibly can, precisely so that when something goes wrong, you can tell which change caused it. Here, nobody had that luxury. The organisation changed and the technology changed in the same cycle, against a fixed statutory deadline that couldn’t move.
What should have accompanied a change of this scale is obvious with the benefit of a Business Continuity lens: a proper parallel run or pilot at meaningful volume before full national rollout, a tested fallback process if the new platform underperformed, and contingency capacity built in from day one rather than added in a scramble once the deadline was already at risk. None of that is exotic. It’s standard practice for any organisation migrating a critical system, and it’s precisely the kind of testing that gets skipped when a contract handover is treated as a procurement and logistics exercise rather than a resilience one.
A supplier risk governance failure
This is also, at its core, a supplier story, and it’s one I keep coming back to with clients. When a critical service moves from one supplier to another, particularly one this consequential, the due diligence doesn’t end once the contract is signed. Ongoing monitoring, agreed performance thresholds, and a genuine understanding of what happens if things go wrong in the early months of a new arrangement all need to be built in before go-live, not discovered afterwards.
The government, as the client here, held ultimate accountability for the resilience of a service that has direct consequences for hundreds of thousands of children and their schools. Whatever assurance process existed before this contract went live, it didn’t catch a platform that markers have described as barely able to cope with demand. That’s not really a story about Pearson’s competence in isolation. It’s a story about whether anyone stress-tested the new arrangement, at scale, before betting a statutory national deadline on it.
A crisis communications failure
The third thread is the one I found most interesting, because when I went looking for a statement, I did eventually find one. Pearson published a statement confirming the delay, admitting the cause was technical issues with the new marking platform and problems transferring data between systems, taking clear ownership of the disruption, and confirming results, originally due on 7 July, would now follow on 16 July. As statements go, it’s actually a reasonably good one. It names the problem, it doesn’t hide behind vague language, and it thanks the markers who’d been dealing with a broken system for weeks.
The trouble is where it was published. It sits in the News and Insights section of Pearson’s corporate investor relations site, in amongst annual results, share price updates and sustainability reporting. That’s a perfectly sensible place to tell shareholders and journalists something happened. It is not remotely where a worried parent, an anxious Year 6 teacher, or a headteacher fielding calls from families would think to look. A good statement that nobody who needs it can find is functionally the same as no statement at all to the audience who actually matters in the moment.
This is a distinction I see trip up organisations more than almost anything else in a live incident. Having a crisis communications plan isn’t just about drafting the right words. It’s about knowing, in advance, where the people affected will actually go looking, and making sure the message is sitting there when they arrive rather than three clicks deep on a page built for an entirely different audience.
Why this matters beyond one exam board
None of these three failures are unique to Pearson, to Sats, or to the education sector. They’re the same three failure points I see across almost every serious incident I’m called in to help unpick: a change introduced without proper testing, a supplier relationship monitored at the point of signature rather than throughout delivery, and a communications response that either doesn’t exist or doesn’t surface where the people affected can actually find it.
The uncomfortable truth is that any of the three on their own is usually recoverable. It’s when they stack, as they appear to have here, that a manageable technical hiccup turns into a national story and a call for the entire model to be dismantled. Worth asking of your own organisation: if a critical supplier changed tomorrow, would the testing catch a failure before it went live, would anyone be watching closely enough in the first few months to catch it early, and if the worst happened, would your crisis statement actually be sitting somewhere the people affected would think to look, or somewhere that only makes sense to shareholders and journalists.
