Last Saturday was the 10 year anniversary of my first day at the Office for National Statistics. This is a happy memory. I had a brilliant boss, wonderful colleagues, supportive leadership from the very top, GDS was in its most impactful and collaborative infancy and I got to work on some really hard problems and do the best work of my life.

To this day it was my most agile, user-centric and data informed team. We definitely made mistakes – and I certainly contributed to that big time with some decisions – but I remain proud of what we did and how we did it. 

It was an exciting time for a few years there. It felt like we were a part of something that was really making a change in the way public service approached technology (and design)…but a decade on and I have to be honest I’ve spent the last week feeling a bit down. How much did we really move the needle? Did any of those changes really stick?

Intellectually I know things have changed and for the better – ten years ago I couldn’t have dreamed of a scenario where there were so many designers, developers, researchers, product and agile people etc throughout every corner of the public sector. That ‘cloud’ would be the default and data would be opened up. Things like Notify, Pay, the Design System etc would have been a dream.

There are pockets of talent all around the country doing work that stands up against anything being done by teams in ‘Big Tech’.

…but all that said we still have stakeholders yearning for the false certainty of a delivery plan filled up with milestones (guesses) who distrust agile believing it lacks rigour. 

We still have teams telling us we don’t need user research because they understand what they really need. 

There is still foundational ‘legacy’ (i.e. bloody old) technology underpinning important public services everywhere that nobody has been able to tackle yet.

Big consultancies have returned to every corner of the public service digital landscape – and the era of encouraging small and medium expert enterprises who would be judged by outcomes seems to have passed…in fact it looks like it is being designed out.

We managed to turn the GDS product lifecycle (discovery/alpha/beta) into a waterfall process with assurance gateways and checklists rather than peer reviews.

Horrible as the peak of the Covid era was, it really did feel like we were finally embracing true lean/agile ways of working and delivering. Things moved fast but teams listened to users, iterated, deployed and learned again. They leaned on open source, shared findings, embraced platforms and got things done.

Since things settled down though it feels like we have retreated. Not just to the time before the pandemic but further back than that. Before Brexit, before GDS was split asunder (there is lots of great work happening at CDDO but does anybody know?) to some unspecified time before any arguments were won.

Forget about Perm Secs, how many Director General level people come from a digital/tech background? A decade plus in really how much influence do we have? Or are we still having to make the case for these ways of working as if we are a supplier?

Or am I just old, grumpy and feeling a bit blue? Did my recent milestone birthday just tip me into old Abe Simpson shouting at a cloud? Are things brighter than I am painting them?

Or is it all a bit Groundhog Day?