Process -The Forgotten Art of Innovation
Everyone wants to talk about the shiny thing. The new AI model. The automation platform. The dashboard that finally makes the data "make sense." Innovation, in most boardrooms, has quietly become a synonym for technology - as if buying the tool is the transformation.
It's not. And the sooner more leaders admit that, the faster their initiatives actually start working.
Here's the truth from the inside of enough carve-outs, integrations, and digital transformations to know it cold: technology rarely fails because it's the wrong technology. It fails because nobody designed the process it's supposed to run on. Hand a team the best CRM, the smartest automation layer, or a frontier AI model, and if the workflow underneath is broken or lives entirely in three people's heads, you haven't transformed anything. You've just built a faster way to produce the same mess.
The Discipline Everyone Skips and Shouldn't
Process design doesn't get a keynote slot. Nobody puts "I mapped the swimlane" on a conference badge. But process is the connective tissue between strategy and execution. It's the difference between an organization that adopts new technology and one that just purchases it.
Here's what actually happens inside a growing company mid-transformation:
Leadership commits to a bold roadmap
IT stands up the tools
Six months later, adoption is patchy, workarounds have multiplied, and the "old way" is still quietly running in a spreadsheet somewhere
That's not a technology gap. That's a process gap. Nobody asked: What is this workflow actually trying to accomplish? Who touches it, in what order, and why? Who's accountable for the decisions along the way? Skip those questions, and technology just automates the dysfunction faster — with better UX.
The Only Metric That Matters: Adoption
Here's the test we'd put in front of any leadership team evaluating a new tool, platform, or AI initiative: the goal of innovation is to make people more effective. Full stop. Not to look advanced. Not to win an award. Not to make the org chart feel modern.
If a new capability doesn't change how someone actually does their job - faster, easier, with fewer errors and more confidence - it hasn't innovated anything yet.
Cool technology that isn't adopted isn't innovation. It's R&D. That's not a bad place to be - R&D is how you learn what's possible. But R&D and innovation are two different stages, and treating one like the other is how transformation budgets quietly evaporate. If a capability stalls at "impressive demo" and never becomes "how we actually work," it hasn't crossed the finish line. And leaving it stalled there is expensive: it burns budget, it burns credibility with the team that watched it fizzle, and it hands the next initiative a harder crowd, because "here we go again" is a tough mindset to undo.
Adoption is the whole game. And adoption is a process problem - not a technology problem.
Why Most AI Initiatives Stall
Nowhere is this playing out more visibly right now than AI. Every leadership team we talk to is under pressure to "do something with AI." Budgets get approved, pilots get stood up - and six months later, adoption is thin and momentum is gone.
Two reasons this happens. Process is the bigger one.
Reason one: there's no process for the AI to run on. A model can summarize, draft, predict, or automate - but it still has to plug into a real workflow. Who reviews its output? Who owns the exceptions? What happens when it's wrong? Skip that scaffolding, and AI becomes one more disconnected tool people have to remember to use, instead of something built into how work already flows. It gets tried once, doesn't fit, gets quietly shelved.
Reason two: the data isn't good enough to trust. Even a well-built process collapses if the AI running inside it is working from inconsistent, incomplete, or unreliable data. People adopt what they trust - and trust evaporates fast the first time a system is confidently wrong because the data underneath was a mess. Clean, governed data isn't a nice-to-have. It's table stakes for adoption.
Both are completely fixable. Neither gets fixed by buying a better model.
This is a huge share of the work we get called in to do. We regularly walk into projects where the technology's already purchased and licensed — but the initiative stalled, because nobody did the process work or the data work that should have happened first. Our job isn't to replace the tool. It's to go back and build the foundation that should've been there on day one - map the real workflow, fix the data, and rebuild on something solid enough to hold weight.
Process as the Enabler, Not the Obstacle
There's a common misconception that process is bureaucracy — the enemy of speed and creativity. It's the opposite. Good process is what makes speed sustainable. It's what lets an organization scale a good idea instead of just having one.
This is the thinking behind Stonehill's approach to post-merger integration, organizational design, and our AI/Automation/Analytics Center of Excellence work: technology is the multiplier, but process is the foundation it multiplies. Skip the foundation, and you're just multiplying chaos faster.
Bringing the Art Back
Rebuilding process discipline isn't about drowning teams in flowcharts for their own sake. It's about asking the harder questions before reaching for the tool:
What outcome are we actually trying to drive?
Where does this break down today, and why?
What does "good" look like, step by step, for the people who live inside this workflow every day?
Get those answers right, and every technology decision after it gets easier — because you're no longer trying to automate ambiguity.
Innovation doesn't start with a tool. It starts with clarity about how work actually gets done. That clarity is process — unglamorous, often invisible when done well, and the forgotten art that separates organizations that actually transform from those that just spend money trying to.
