About four years ago, my CEO and I attended an intimate conference focused on revenue cycle improvement. Small room, maybe 40 people, senior operators and executives all wrestling with the same challenge: how do you get more value out of your revenue processes? It was exactly the kind of setting where real conversations happen, away from the noise of large industry events.
At one point, the facilitator asked the room where to start when looking to automate revenue cycle processes. The answers came quickly. Automate prior authorizations. Automate claims submission. Automate eligibility verification. Automate denial management. Every answer in the room was a variation of the same thing: pick a process and put automation on top of it.
When it was my turn, I gave a different kind of answer.
Before you decide what to automate, you need to decide what to eliminate. Then simplify what survives. Only after that does acceleration, through technology and AI, actually pay off.
The room went quiet. Not because the idea was radical, but because nobody had framed the question that way. Everyone was so focused on where to apply automation that they had skipped two steps that determine whether the automation creates value or just moves problems faster.
That conversation became the foundation of a framework I have used in every organization since. I call it E-S-A. Eliminate, Simplify, Accelerate. In that order, always.
Start with the customer, not the process
The instinct most organizations follow when looking to improve operations is to map existing processes and ask where technology could make them faster. That instinct is understandable, and it is almost always wrong.
The right starting question is: what does the customer actually value? Not what your organization delivers, but what the customer needs to receive to do their job, make their decision, or feel confident in the relationship. Everything that does not connect to that answer is a candidate for elimination, not optimization, not digitization; elimination.
This matters because most organizations are carrying a significant amount of work that exists only to support other work, rather than to serve the customer directly. Approval layers that exist because someone did not trust a process years ago. Reconciliation steps that exist because two systems never got properly integrated. Manual handoffs that exist because nobody redesigned the flow after an acquisition or a reorg. None of that work has value to your customer. And none of it gets better by accelerating it.
The discipline here is intellectual honesty. It requires a leader who can look at operations end to end, across functions and systems, without being anchored to how any one department has always done things. That is exactly the mindset a great CDIO brings to this work. Not a technology operator, but an executive who sees the full value chain and has both the standing and the skill to challenge what does not belong. That conversation is rarely comfortable. But it is the one that creates the most value, often before a single technology investment is made.
Simplify before you systematize
Once you have eliminated the work that should not exist, you are left with the work that genuinely needs to happen. The question now is whether that work is as simple as it can be.
Complexity is one of the most expensive things a business carries, and one of the least visible on a balance sheet. It is expensive to maintain. It is expensive to train people on. It is expensive to fix when it breaks. And it is very difficult to hide from your customers, even when you try hard to do so.
Customers experience your internal complexity as friction in their relationship with you. When a customer asks why a simple change to their quote takes four weeks, they are not interested in your approval workflow. They are telling you that your complexity is costing them time and eroding their confidence. When a distributor cannot get a clear answer on when their order will arrive, that is your internal coordination problem showing up in their operations. When a patient stands at a pharmacy counter with no idea what they owe, that is your billing complexity making them feel like they are dealing with three different organizations instead of one.
Friction erodes trust. And in a B2B environment, trust is the foundation of every renewal, every expansion, and every referral your business depends on.
The goal of simplification is to make the right path the easy path, and to make exceptions genuinely exceptional. Standardize the decisions that follow predictable patterns. Remove the steps that exist only because the underlying process has never been cleaned up. Give your people clarity, and your customers will feel it.
When you simplify first, you are not just reducing cost. You are building something that is worth accelerating.
Now you are ready for AI
This is where the business case for E-S-A becomes even more compelling, because the technology available today is categorically more powerful than what existed even three years ago.
AI can handle judgment, not just repetition. It can work with unstructured information, surface patterns across large data sets, and make recommendations that previously required experienced human analysis. The opportunity is real, and the leaders who capture it will build durable competitive advantages.
But AI does not fix a broken process. It scales it.
I saw this play out firsthand at a diagnostics business where customers needed a steady supply of consumables to perform their work. For years, the process was manual. Someone would estimate how much to ship based on experience and intuition, and they were wrong often enough that customers would run short mid-month. When a diagnostics lab runs out of supplies, they cannot serve their patients. That is not an operational inconvenience. That is lost revenue for the customer, and a damaged relationship for us.
We applied E-S-A to the problem. We eliminated the guesswork entirely. We simplified the data inputs so usage patterns were visible and clean. Then we built automated forecasting that predicted what each customer needed and shipped it before they had to ask. The result was straightforward: customers almost always had what they needed, when they needed it. Lost revenue events dropped dramatically. The relationship shifted from reactive to proactive, which is a very different commercial conversation.
That outcome was not possible by automating the old manual process. It required redesigning the process first.
If you hand a cluttered, redundant workflow to an AI tool, you do not get a better workflow. You get a faster, more consistent version of a bad one, and that version is often harder to fix because it now feels systematic and permanent. The dysfunction has been encoded.
The E-S-A sequence exists precisely to prevent that outcome. When you eliminate waste and simplify what remains before you bring AI into the picture, you give the technology the conditions it needs to create real value. Clean inputs. Clear logic. Predictable decision points. That is the environment where AI moves from pilot project to business driver.
And the returns compound. A workflow that has been through E-S-A and then enhanced with AI does not just perform better today. It becomes easier to improve over time, easier to scale as the business grows, and easier to explain to your board and your customers. You are not managing a layer of tools bolted onto legacy processes. You are running something you intentionally designed.
The cost of skipping the sequence
The pressure to show AI progress is coming from every direction right now. Boards are asking about it. Investors are asking about it. Customers are asking about it. The temptation to launch a pilot, announce a deployment, and move on is real.
But dropping an AI tool on top of an existing workflow does not simplify anything. It adds a layer. That layer needs to be integrated, monitored, governed, and explained when something goes wrong. Complexity that was already expensive becomes more expensive, and the underlying problems that created the friction in the first place are still there, now harder to see and harder to address.
The leaders who will win with AI are not the ones who adopt it fastest. They are the ones who prepare the ground first, so that when AI is applied, it accelerates something that was already working cleanly.
Invest early in reducing complexity and you get two returns. The short-term return is a simpler, faster operation that your customers notice immediately. The long-term return is a foundation that AI can genuinely build on, one that appreciates in value every year as the technology improves. Skip that investment, and every AI dollar you spend is accelerating debt, not creating value.
The sequence is the strategy
E-S-A is not a technology framework. It is a business discipline for leaders who want technology to create lasting value rather than recurring cost.
Eliminate what the customer does not value. Simplify what remains until the process is as clean as it can be. Then accelerate with AI and technology to make it faster, smarter, and more scalable than your competitors can match.
The organizations that follow this sequence will not just get more from their technology investments. They will build the kind of operational foundation that compounds in value over time, attracts better talent, and earns the kind of customer trust that is very hard for a competitor to take away.
Get the sequence right. Everything else follows.