Carmel Partners

Toward More Effective Technology Transformation

There has been a cyclical pattern to technology efforts that moved from buying individual best-of-breed tools to focusing on larger tools to consolidate and simplify environments. This created a constant state of change management and always came with trade-offs, no matter which way you were going.

Now, the idea of the “platform” is the long-term strategy. This isn’t just one mega-application that does everything. It is about systems that are integrated, not just interfaced. We can incorporate the newest tools with the best features, but they must fully integrate with the existing structure. We need the systems to share security and access controls, consolidated health monitoring, and most importantly, data.

When this works, it improves adoption with your user base and reduces your tech debt. It also does away with the old risk of relying on that one person who would run a weekly report in one system and manually type the data into another. Like building a puzzle, the pieces need to fit together and make the same picture.

AI That is Disrupting Transformation Today

You can take that old three-year roadmap and throw it out the window. We need to be much more flexible now with AI and its speed of change. Your roadmap becomes a picture of how the business evolves toward a collective North Star. This helps prioritize truly meaningful projects and not just process enhancements. If you don’t have a picture of where you want to go, agentic AI sprawl can quickly become a problem.

That flexibility isn't just about scope. It's about how we fund it, too. There are fewer known licensing costs and more usagebased expenses. Budgets are going to need to adapt to this. Justifying projects with simple cost-offset comparisons won’t work when the actual spend is changing every month. We must show AI’s value in terms of business impact, not cost offset.

Challenges of Integrating AI into Existing Technology Environments

Three words—governance, sprawl and data.

With governance, understanding the current and upcoming regulations around AI falls on you. And since these are also constantly changing, you are often just answering the question, “Can I justify this to a regulator?” The lack of strong administrative controls in many AI tools makes managing them difficult, and so you may have to look at how to defend the rest of the platform from the AI instead.

With the ease of building applications with AI, application sprawl and shadow AI are high-priority challenges. Even with well-structured and controlled environments, it is easy to turn around and find someone has built a web application, and the AI has found a place to post it.

Lastly, your data has to be well-structured and clean before you implement AI. You used to have a group of business intelligence professionals or data scientists who could interpret the intricacies of a complex database and know how to decipher the labels for data elements. Now, end users across the company are asking the AI for information and expecting it to understand the data in the same way.

Without data governance to determine what data the AI should be reading and how it should interpret it, you run the risk of hallucinations and inconsistency in the AI responses.

Balancing Innovation with Security, Scalability and Business Continuity

You can’t be seen as hindering innovation. You have to get governance and security in place as fast as possible so you can enable safely. Structure the plan around how to give what you can now, and how you can phase in more down the road.

“ We can incorporate the newest tools with the best features, but they must fully integrate with the existing structure. Like building a puzzle, the pieces need to fit together and make the same picture. “

Building a review board that includes technology, data science, compliance, and even HR will help you find allies and open the discussion of risk and where your company’s tolerance is. Be transparent about the “why” for any rules you put into place so there is understanding of what you are doing. You need to be seen as a defender of the company, not a roadblock.

Navigating AI-driven Transformation

Work with leadership to get the “North Star” defined. Create the vision for how the company is going to truly evolve with these new tools and determine how you are going to measure that success. This will be your guidance on decisions around the controls you need to have in place, the tools you implement, training, and more.

I would also encourage you to build your AI community within the company. The teams that trust and adopt AI will also be the ones who uncover new ways to use it that may be the spark that changes your business.