Working with Data...Was KES 12,500
Back to blogsOriginal post
You Don't Need Another Tool. You Need a System.
SystemsData StrategyAIWorkflow

You Don't Need Another Tool. You Need a System.

By Ezra Muinde on Apr 23, 2026 - 5 min read

Ubuntu Analytiq insight

You can have Python as your striker, Excel in midfield, even AI like ChatGPT on VAR, but without a system, you are not winning the league.

Here is the uncomfortable thing nobody says at the data conference. The CEO does not care what you used.

Not Python or Java. Not Claude or ChatGPT. Not Power BI or a hand drawn chart on a whiteboard. What she cares about is whether the insight changed the decision. Full stop.

Yet most professionals in data and AI spend most of their energy learning the newest tool. The latest library. The hottest model. The framework that promises to write code on their behalf. Then the next one. Then the one after that. Always starting, rarely finishing, never resting.

The problem is not the tools. The problem is that we are confusing tools for systems. And they are not the same thing.

Tools expire. Systems evolve. One requires you. The other replaces you.

Systems Keep Running

There is a theological idea worth sitting with here. When God finished creating He rested. Not because the work ended but because the system was complete.

Think about that. He created trees and never created another tree again. The seed produces the tree. The tree produces the fruit. The fruit falls and produces more trees. The system runs.

Whether you plant on loam soil, sandy soil, or clay. Whether you are doing hydroponics or running a greenhouse. The environment changes. The tools change. The conditions change.

But the system does not.

The seed still becomes the tree. The tree still produces fruit. The cycle continues.

Now bring that into Egypt. Joseph did not create food. He built a system. Seven years of surplus stored then distributed through famine. The food was the tool. The system made it matter.

That is the difference.

The Arsenal Explanation

We are told we were created in His image. If that is true then maybe the goal was never to master tools. It was to build systems that work even when we are not there.

Now here is where it gets practical. And I will explain like Arsenal so you do not lose your entire bet slip and start questioning life decisions.

You can have the best striker which is Python. The best midfield which is Excel. Even VAR on your side which is AI tools like ChatGPT or Claude. But if there is no system, no formation, no structure, no understanding of how the team plays, you will still lose the match. You will not even get one cup leave alone a quadruple.

A VLOOKUP in Excel is just a join. The same thing you do in Python with pandas. The same thing you model in Power BI. The same concept in R. Different syntax, same system.

So the real question is not what tool are you using. The real question is what are you trying to achieve.

Input, Process, Output, Feedback

Excel has been around since 1985. Python since 1991. Both are still here. Not because they defeated newer tools but because the system they serve has not changed.

Input. Process. Output. Feedback.

That loop is eternal. Excel serves it. Python serves it. AI serves it. The tools are interchangeable. The system is permanent.

I have used many tools across my career. Excel was my first analytics tool and I still use it. Then SPSS. Correlations, regressions, cross tabs, frequency tables. Then I taught myself R, did not like it, and ended up translating most of it into Python anyway because Python is what I love. That is what I analyze with.

Jupyter Notebook inside Visual Studio Code. Dark mode gang, we do not negotiate. Google Colab. Then my organization said we cannot use public LLMs, so I switched to GitHub Copilot after they standardized on it. Now I am even building Copilot agents because adaptation is survival.

And here is the thing I learned through all of it. When I moved from tool to tool I was not switching systems. I was upgrading the interface. The thinking underneath never changed.

Upgrades, Not Restarts

If your thinking is tool based you will keep restarting every time something new shows up. If your thinking is system based every new tool becomes an upgrade not a reset.

A few months ago there was hype around Prompt Engineering as a career. I checked recently, silence. Now it is '7 prompts to make your ChatGPT more productive'. The title changed. The system did not.

70% of organizational transformation projects fail. Not because of lack of technology but because of poor strategy, weak processes, and absence of systems thinking. McKinsey has said this clearly.

Not lack of AI. Not lack of Python. Not lack of dashboards. The tools were present. The systems were not.

And here is another one. Organizations that consistently document, monitor, and improve their processes are about four times more likely to outperform those that do not. Four times. Not because they had better tools. Because they had better systems.

Build the System

Think about the organizations that are always reacting instead of planning. No data system so they guess. No reporting system so they explain everything from scratch. No feedback loop so mistakes repeat. Then they hire more people and still do not grow.

That is not a people problem. That is a systems problem. You cannot scale by stacking humans on top of a broken process.

The most valuable professionals are not the ones who know the most tools. They are the ones who understand workflows, structure problems clearly, build repeatable processes, and think in loops not tasks.

Maybe the reason you cannot rest is because you do not have a system. You have a stack of tools.

When God finished creating He rested. Not because there was nothing left to do. Because what He built no longer needed Him to run it.

We are told we were made in His image. If that is true then maybe the highest form of work is not mastering every tool that arrives. It is building something that keeps working long after you have stepped away from it.

So stop asking what tool should I learn next.

Start asking what system am I building.

Because systems do not just produce results. They produce consistency. They produce clarity. And eventually if you build them well enough they produce rest.

Next step

Turn the idea into practice through the Academy, Mentorship, Projects, or Datasets workflow.