The Green Superman
Another incredibly busy week at work, with plenty of socialising too. It’s actually been really fun and good progress is being made in multiple areas, so I’m feeling happy.
Extract
This week we iterated the prototype a lot, based on feedback from users in the first round of research. I’m not going to go into everything here, you can check out the latest weeknote for Extract for that – but beware, it’s a juicy one.
There’s been some planning ahead to the beta too, as the spring 2026 deadline set by the Prime Minister comes closer into view. The main thing I need to do is line up local authorities to be involved in the testing, and also help the department line up my replacement. I’ll need them to shadow me as we plan the beta, so that that can act as the handover.
Splitting product and technology
As we’ve been reviewing insights from research with users, I’ve been keen to split which are related to the product space and which are related to the technology. i.AI are very eager to prove that they’ve incubated a technology to replace a manual process, and from an R&D perspective they’ve definitely done it. But in order to shape this solution into the right product for the users and their problems, we need to go deep on the research, surfacing conceptual, contextual and usability issues that may inhibit the product’s success.
As I’ve said in previous weeknotes, innovating a technical solution is not the same as designing a product. You can pivot technical solutions into new problem domains simple by giving them a different product wrapper.
Take the Segway, for example. It was once lauded as a revolution in the transport space, and people said it would change everything. The innovative technology, using tilt sensors, gyroscopes and motors, meant you could lean forward or backward to move the Segway. It’s fast, uses an electric battery, and is small enough to store in a large cupboard.
But it never disrupted the car or public transport. Segways quickly became novelty products, with more being bought by city tour operators than everyday people. The car was simply a better product for transportation, and Segway never cracked it.
Remember hoverboards a few years ago? Those use the same technology as a Segway, just in a smaller, more convenient, and cheaper package. They’re really popular around me in south London, I still see people riding around on them. But I haven’t seen a Segway in years.
The ability to pivot – and I mean truly pivot – defines the split between product and technology too. Before developing Figma, Dylan Field and Evan Wallace were making bets on WebGL as a technology, coming up with products that utilised the technology. Early ideas included a meme generator, something to do with drones and I’ve seen mention of slide deck software too. It wasn’t until they decided to create ‘Photoshop in the browser’ that Figma became the product that we know today. That pivot in the product space – while still utilising the same technology – shows how the two concepts can be split, suggesting alternative strategic routes.
That’s why I want to separate the two, rather than think about them as the same thing. Product does not equal the technology, and vice versa.
Skipping the basics
A couple of weeks ago, I realised I’d skipped one of the basics of setting up a team: establishing roles and responsibilities. Calling myself out on this because it’s a silly mistake, especially because I have a list of must-haves when starting or joining a product team. Facepalm.
Anyway, on reviewing the list I decided to add a couple more reminders to Future Steve, for the next time I set up a team. Hopefully you’ll find them useful too.
Garden notes
Not much garden action this week, and especially not this weekend. It has been windy and rainy!
Some broad beans I sowed last weekend are yet to pop. I’m worried the soil I used was too waterlogged, or perhaps it’s been too warm in the house. We’ll see, they’re not supposed to pop for another couple of days but I’m still a bit anxious.
Bookmarks
- Small acts of maintenance, 3 mins. Really lovely piece from Nia Campbell. The idea of ‘organisational inheritance’ is a good one, and in line with something I often quote from Ellie: ‘We should do the hard work to make [things] simpler for those who come after us.’
- Taking Jaggedness Seriously, 26 mins. Great piece on making real-world adoption of AI happen. Leaning in to the jagged frontier requires you to listen to users and focus on human-computer dynamics. (I referenced this on the latest weeknote for Extract too.)
- Everything I Got Wrong About Product (So You Don’t Have To), 6 mins. I nodded a lot when reading this, and I’ve learned these same things over the years. So glad someone wrote it down!
- Great (public) expectations, 17 mins
- The AI boom is based on a fundamental mistake, 2 mins
- This isn’t a startup, 2 mins
- Why Off-the-Shelf AI Tools Fall Short in Planning, 5 mins
- Online planning is 30 years old today., 14 mins
- On writing…, 4 mins
- 2 Big Questions for AI Progress in 2025-2026, 16 mins
- Rachel Reeves and the Daily Mailification of the BBC, 5 mins
- Could 2026 be the year of public interest technology?, 8 mins
- The guts of the machine, 5 mins
