There’s a persistent myth in enterprise transformation that I think slows more companies down than any technology gap ever could.
It’s the idea that if you want to digitize work, automate processes, or scale analytics, you first need to centralize everything in the hands of technical specialists.
That sounds disciplined. In practice, it often creates distance between the people who understand the work and the people expected to improve it.
If you want to build a truly intelligent enterprise, you have to do more than add technology to existing workflows. You have to redesign how improvement happens. That means giving more people the ability to solve problems directly, especially the people with the deepest domain expertise. In many cases, the person who best understands the bottleneck is not in IT. It’s the person living with the bottleneck every day.
That has been one of the most important lessons in our journey at Siemens Energy.
Rethinking who gets to build
The moment our thinking shifted was when we stopped asking, “How do we get more technical resources?” and started asking, “How do we get the right data into the hands of the people who understand the business and its processes?”
The first question assumes a process where analytics stay centralized, and insights flow down from a team of specialists. The second question assumes that the people who understand day-to-day business problems — the procurement manager who knows what a supplier delay costs, the factory analyst who understands why a production metric is off — are the ones best positioned to build solutions.
We call this approach Citizen Development. The core idea is simple: enable people who are not traditional coders to build analytics solutions using low-code tools. Give them access to data, the right platform, and some support from technology specialists and great things will come up.
Building the foundation for data access
Before we could empower our citizen developers, we had to solve the “access to data” problem. If people need to file a request or export a massive report just to see raw data, the friction is too high. Nobody builds anything. They go back to their spreadsheets.
So, we built the SE Data Center — and Alteryx macros which connect to Snowflake where raw SAP data of all our various SAP systems is replicated every 20 minutes. This forms the core of our data democratization concept.
We included guardrails by building standardized workflows, reusable templates, and a centralized parameter register — giving teams the freedom to build while maintaining data integrity. That balance between independence and structure is something I’d encourage any organization to think hard about before they scale.
What scaling self-service analytics looks like
We started in 2018 with just a few licenses to explore, but we quickly gained momentum. By 2020, we had already reached 100 Alteryx licenses and were using them for a variety of use cases. The first major project was automating a weekly cash-in report — not glamorous, but it worked, and it freed up hours that analysts had been spending manually.
Word of the win spread to the procurement domain, where one of our operational buyers was spending several hours every week manually drafting emails to suppliers about pending purchase order confirmations. We built a workflow that analyzed our factory’s raw SAP data, identified every missing PO item confirmation, and automatically sent emails to our suppliers in various languages.
The cycle kept going as one win inspired another, which eventually led us to the Procurement Cockpit — an end-to-end visualization suite powered by Snowflake, Alteryx, and Tableau that gives operational buyers and procurement teams real-time transparency across the entire operational procurement process chain from demand to goods receipt.
Today it runs across more than 20 factories in nine countries and saves over 150,000 hours annually. That single solution, especially the underlying dynamic workflow architecture, became a blueprint that we can adapt and replicate in other highly scalable, governed use cases.
Domain expertise is becoming more important in the age of AI
With all the attention on AI, there is a temptation to assume the future will rely less on analysts closest to the work.
I see it differently.
As automation and AI lower the barrier to building, configuring, and scaling solutions, domain expertise becomes even more valuable. Not less.
Why? Because the hard part is often not building the workflow. The hard part is knowing how the data is connected to each other through the process, what the workflow should do, what exceptions matter, what decisions need support, what business logic should govern the process, and what an outcome actually means in context.
That understanding comes from people who know the business and its processes. It also builds the foundation for AI use cases because raw data is enriched with business logic.
This is one additional reason low-code platforms and AI can be so powerful. They do not just accelerate technical execution. They allow more people to participate in improvement and transfer a business question into a data driven solution.
The organizations that get the most value from these technologies will be the ones that connect them to the judgment of the people closest to the work. The Siemens Energy journey makes this clear: the push toward citizen development, broader participation, and experimentation was inseparable from the company’s progress in analytics and automation.
How leaders can get started
For leaders, the challenge in becoming an intelligent enterprise is whether your organization is building the right conditions for that shift to happen in a way that is scalable, trusted, and connected to the people who understand the work best.
I recommend these practical steps to get started:
- Start with a real problem, not a proof of concept
- Bring end users to the table on day one
- Treat data access as a prerequisite
- Put governance and guardrails in place early
- Design for reuse, not just resolution
The companies that will define the intelligent enterprise will be the ones that figure out how to put analytical capability in the hands of the people who understand the work — and then built the culture, the infrastructure, and the trust to let those people use it.