Last week, with only one week’s notice, 52 People leaders signed up to spend a morning with us talking about how to build a high-performing, AI-enabled organisation.
I think that tells you how much pressure people are feeling right now to "do something about AI" while still being expected to lift performance, hold culture together, and navigate more change than most of us have seen in our careers.
But AI is just the catalyst for a conversation we needed to have anyway. We've been circling it for years.
Across the panel and the breakout rooms, the same three tensions came up again and again:
🧭 People don't know how to get started with AI adoption, let alone lead it
😮💨 The pressure to build high-performing teams while uncertainty grows is overwhelming
🐌 The traditional HR approach is too slow to keep up
If any of that feels familiar, this one's for you. This week’s newsletter is a summary of the roundtable and a conversation that Mark and I had the following day to reflect on the learnings.
PREVIOUS NEWSLETTERS
In case you're a new reader (or just missed them), here are the past few newsletters:
- Would you rather follow a script - where we share why high-autonomy cultures often need just as much manager support as low-autonomy cultures.
- Accountability isn't an HR problem - where we challenge the assumption that measuring platform usage or adoption equals accountability for performance
- The choice to co-create that changed everything - where we showcase Pixelogic's iterative approach to building company-wide ownership for performance
Why our definitions of HR, AI and performance get in the way
Before you can build a high-performing, AI-enabled organisation, you have to be honest about whether you’ve actually agreed what any of those words mean.
One of our panellists pointed out that most organisations have a shaky definition of performance before AI even enters the picture, so when leaders rush to encourage AI adoption to drive performance, they're trying to improve something they were never measuring well to begin with.
When I unpacked this with Mark the next day, he said it’s important for HR folks to understand that what a CEO cares about most is the business moving forward. So, most of them will assess high performance as someone’s ability to help the organisation move from A to B efficiently. And in this day and age, AI is seen to be driving that efficiency.
(One word of caution, though: don't confuse volume of AI usage with performance. Measure the wrong thing, and people will game it every time.)
Then there's the term “AI-enabled” - Access to tools isn't the same as capability. Having Excel doesn't make you financially literate, and handing everyone a chatbot licence doesn't make an organisation AI-enabled. The work is designing your organisation so technology does what it's best at, people do what they're best at, and the two together create more value than either could alone.
Which brings us to HR's own shift, the one the whole panel agreed on: performance is something you design, not something you manage. Create the conditions for people to succeed rather than pushing top-down initiatives and hoping they stick.
How to take a People Experience approach AI adoption and performance
Here are three suggestions from our panellists on how to get started with building a high-performing, AI-enabled organisation using a People Experience (aka a product management & human-centred design approach).
🔍 Start with discovery, but keep it small. Resist the urge to jump straight to a tool; instead start by analysing where the value is being lost. Ask about where we are:
- spending time and effort that human-skills don’t create value and preventing people from doing their best work?
- wasting time searching for information that we’re struggling to access?
- experiencing unnecessary complexity, frustration, or difficulty in our processes for our customers AND our people?
Keep discoveries small and targeted on one real friction point, not a sprawling six-month diagnostic.
🔄 Iterate instead of trying to get it perfect. I’ll admit something. I used to be a discovery purist. Discovery first, always. But that was largely because we didn’t have good ways to experiment, so a big upfront discovery felt like the only safe way to make a decision. I’ve changed my mind. It’s far easier to build buy-in when you can gather data as you go and move the problem forward instead of going in circles trying to define it.
Gather enough data to figure out which work is highest value (worth holding onto) and which is lowest value (ripe to redesign or hand to AI), and then start testing.
🫱🏽🫲🏾 Co-create, don’t design for people. The people closest to the friction usually have the best ideas for fixing it. Shift your perception of your role from being the expert with the answers to being the person who helps the team surface opportunities for AI enablement and act on them. Our role in HR should be to encourage teams to build reflective practice into their ways of working, enabling them to spot frustrations in their workflows and experiment with AI to free up their time for more meaningful work.
Where to get started
The day after the roundtable, I caught up with Mark to compare reflections, and a few things crystallised about where to actually begin. Here are the three we'd point you to.
1. Define what performance means in your organisation and how AI enables it.
2. Start by improving one painful process with AI; don't rebuild everything - leave room to experiment and learn from the changes
3. Visualise end-to-end processes and capture data on where they're breaking so you can see what to refine and iterate (hint: Crewmojo helps with exactly this - book a demo to learn more).
The organisations that pull ahead won't be the ones that adopt AI fastest. They'll be the ones that use it as a reason to finally design work around people, on purpose.
If you're wrestling with any of this, get in touch; we'd love to help.
Subscribe to receive the latest posts to your inbox every month.