Been around this tech world for a good while now, over forty years if I’m counting right. The last decade or so, mostly focused on acquisitions at Constellation, has offered a different kind of education. You see, when you’re looking at dozens, sometimes hundreds, of niche software businesses a year, you start to notice patterns. But more importantly, you start to appreciate the absence of a single, universal pattern for success.

I was chatting with a young, bright BU manager the other day – we’d acquired her company about three years back. She was a bit frustrated. She’d heard about a really clever customer retention strategy another BU, in a completely different vertical, was having success with. She’d tried to replicate it, meticulously, but it just wasn’t clicking in her market. "I don't get it, John," she said, "It worked for them. Why not for me?"

It got me thinking. In many large organizations, the instinct is to find the "best practice" and then roll it out with a mandate from on high. Someone smart in a central office – and there are plenty of smart people – devises a grand plan, and everyone is expected to fall in line. The thinking is, "We know this works, so let's standardize for efficiency." And sometimes, for very specific, mechanistic processes, that makes sense.

But for most things that really move the needle – how you approach a market, how you structure a new service offering, how you engage a particular type of customer – context is king. And we, at the centre, often don't have the deepest context for each of our hundreds of verticals. Our BU leaders do.

What we have at Constellation, almost by happy accident in the early days and now more by design, is a vast, distributed laboratory. Each business unit is, in its own way, running experiments every day. They’re small, mostly. A new pricing tier for a niche product. A different approach to a sales demo for a particularly stubborn client segment. A tweak to their support model.

Most of these experiments, frankly, probably don't yield earth-shattering results. Some fail quietly. And that’s the beauty of it. The "blast radius," as Jeff Bender might say if he were talking about something less benign, is tiny. If a BU tries a new marketing campaign and it flops, it’s a learning experience for that unit, maybe a few thousand dollars and some time. It doesn’t derail the whole group.

But when something does work? That’s where the quiet elegance comes in. There’s no memo from head office saying, "Everyone must now adopt the 'Wombat CRM Integration Tactic!'" Instead, word gets around. BU managers talk. They share war stories at our informal get-togethers. Someone mentions, "Hey, we tried this thing with our small municipal clients, and our churn dropped by X%." Another manager, perhaps in a completely unrelated field like, say, software for private golf clubs, might prick up their ears. Not because they were told to, but because they’re curious and they’re always looking for an edge.

They’ll reach out. "Tell me more about that, Sarah." And Sarah will. She’ll share what worked, what didn’t, what the pitfalls were. The golf club software manager might take that idea, adapt it, twist it for their own unique market. Maybe it works for them too, or maybe it leads them to a different, equally valuable insight.

We had a situation a few years back where one of our businesses in the public transit space was experimenting with a predictive maintenance module. It was a bit of a gamble, took a couple of years to really get right. For a while, it looked like it might not pay off. But they stuck with it, listened to their early adopter clients. Eventually, it became a significant differentiator for them. Other BUs, some in fleet management for completely different industries, heard about it. Not through a formal "innovation summit," but through the grapevine. They saw the underlying principles, not just the specific application, and a few started their own, tailored experiments.

The point is, we at the centre don't need to know the "one right way." In fact, embracing the idea that we don't know is a tremendous asset. It forces us to rely on the collective intelligence and the diverse, on-the-ground experience of our many teams. It allows for a kind of natural selection of ideas. Good ones tend to spread, not through edict, but through demonstrated success and organic sharing. Bad ones tend to wither on the vine without causing systemic damage.

This approach requires a certain humility. It means trusting our people, deeply. It means accepting that the best ideas won't always originate from the top, or even from the most obvious places. It means our main role at the corporate level, beyond allocating capital, is often just to be good connectors – to make sure Sarah from "Acme Utility Billing" can easily find and talk to Tom from "Beta Marine Software" if they’ve got a shared problem or a glimmer of a shared solution.

So, when that young BU manager asked why the other strategy didn't work for her, my answer was simple: "Maybe it wasn't your experiment to run, or maybe it wasn't the right time, or maybe your market just needs a different tweak. Keep trying things. Keep listening. And keep talking to the other BUs. Someone, somewhere, is probably figuring out something you can use, or you’re figuring out something they can."

There's an elegance to it, this ecosystem of trial and error. It’s not always tidy, but it’s resilient, and it’s constantly learning. And for a company that wants to be around for the long haul, that capacity to learn and adapt, driven by many independent minds, is probably the most valuable asset we have. It’s certainly more valuable than any single "right" answer we could ever dream up in isolation.