The world of data analysis is changing right before our eyes at a dizzying pace. Every other moment, we hear about new artificial intelligence models, further breakthroughs in automation, and tools that are supposedly designed to completely replace human labor. Many people today are wondering: what exactly does a modern data analyst need to know? Is it enough to simply master AI prompts? Or maybe those classic, hard technologies, like good old Excel or reliable SQL, will still be useful? And what about soft skills—do they still matter in a world dominated by algorithms?
My name is Kajo Rudziński, and I have been working with data professionally for over seven years, including my time as a Data Architect, before fully dedicating myself to growing KajoData in 2026. Over the years, I have seen many trends that were supposed to revolutionize the industry. From the perspective of these experiences, I can tell you one thing: to be a truly exceptional analyst today, you have to combine fire and water.
I have prepared a breakdown of ten absolutely crucial competencies for 2026. We will split this into five technical skills and five soft skills. You might be surprised, but one simply cannot exist without the other today. Let’s get started.
1. Excel (Technical Skill)
We are starting with a tool that evokes a smile of pity among many modern analysts—Excel. I know that in the technological world, and especially in the bubble associated with artificial intelligence, people are incredibly surprised that anyone still uses spreadsheets. We assume that since we have massive data warehouses, scripts, and advanced dashboards, everything can be replaced. In theory, that is true. In practice, the truth looks completely different.
Excel is still absolutely critical, and for two main reasons. First, business continuously operates within it. Managers, directors, and operational or financial department employees send each other Excel files daily, calculating commissions, budgets, and plans in them. We, as analysts, must speak to these people in their language. We have to be able to open their files, understand their logic, and potentially transfer those data into our systems efficiently.
Second, for an advanced data analyst, Excel is simply a fantastic scratchpad. When you have a new idea or want to quickly see how tables will interact with each other, nothing replaces the ability to instantly sketch a model in Excel. Before you write complex code in Python, it is often much faster to throw a data sample into a spreadsheet, create a pivot table, and see if your hypothesis makes any sense at all.
2. Asking Questions (Soft Skill)
We are jumping to the first soft skill, which is asking questions. It seems trivial because everyone knows how to put a question mark at the end of a sentence. However, I have seen truly outstanding data analysts in action, and what made them experts was the flawless quality of their questions and their ability to dig deeper.
How does this look in practice? Imagine someone from the board saying during a meeting: “Our sales dropped.” An average analyst goes back to their desk and makes a chart showing the decline. A good analyst immediately probes the topic. Since when did they drop? In exactly which customer segment? By what percentage compared to last year? What does this trend look like on a weekly basis?
Deepening questions allows you to precisely define the problem, which in turn leads to finding a much better solution. When you create a dashboard, you are actually creating a picture, a snapshot of the current business situation. You need to know perfectly well what problem this picture is meant to solve. This skill is very hard to train dry in courses; it comes with experience.
Tip: The best way to practice asking questions is while working on your own portfolio. Download ready-made datasets from the web and bombard them with questions. Although AI can generate a list of great supplementary questions for you, remember that during a live board meeting, you won’t fire up ChatGPT to ask it what you should ask. You must train this muscle yourself.
3. SQL (Technical Skill)
Back to technology. If I had to point to the most important hard technology that you must befriend for life, it would definitely be SQL. Regardless of how frameworks change, how cloud tools evolve, and what innovations AI brings, SQL remains the absolute core language of communication with data.
The vast majority of data that businesses rely on resides in relational databases. In short, this means that tables have relationships with one another. When there are three hundred and twenty-seven of these tables, the relationships become immensely complex. You must know how to manage them optimally.
Of course, artificial intelligence can write or optimize code for us. But to use this effectively, you need to know what to optimize. You cannot trust algorithms blindly. You simply must be very good at SQL. This core must be memorized inside out. Complex SELECT statements, Window Functions, CTEs (Common Table Expressions), multi-level subqueries—there is no leeway here.
4. Understanding Business (Soft Skill)
This is the moment in an analyst’s career that acts as the ultimate filter. Understanding business is a difficult and complex skill. It means that when working in a specific company or industry, you must deeply comprehend the mechanisms of how it operates. If you can only write a formula calculating the difference between revenue and costs, you are essentially a coder, not an analyst. A true analyst understands why that formula was created and what the ultimate business goal is.
Understanding the domain you operate in—whether it’s e-commerce, logistics, healthcare, or B2B sales—gives you a massive advantage. Every business has its “backend” and “frontend.” People who have worked in gastronomy know this perfectly—a beautiful dining room for guests is one thing, and the chaos in the kitchen during rush hours is another.
To give the board real insights, you must know what the company actually makes money on. Take cinemas as an example. It might seem that cinemas make astronomical amounts of money on tickets for the latest blockbusters. In reality, film licenses are so expensive that screenings themselves are sometimes barely profitable. The true drivers of a cinema’s profits are popcorn, nachos, and drinks. A cinema is basically a high-priced snack sales point that uses movies as a lure. With this knowledge, you analyze their data completely differently and recommend different optimization actions.
5. Data Visualization (Technical Skill)
After the data extraction stage, it needs to be shown to the world. In the visualization category, there is currently one king—Power BI. A while ago, I would have put an equal sign between Power BI and Tableau, but the market dominance of Microsoft’s tool has become overwhelming.
There is a trap hidden here, though. If you only know Power BI, you will be competing in the market with a multitude of people who also only know this single tool. Therefore, it is worth mastering alternatives like Looker or Tableau, at least superficially. This will allow you to internalize general knowledge about how good data visualizations are created in the first place, regardless of the technological overlay. Technology changes fast. The tool that is king today might be in retreat in five years.
Your goal is not to be a consultant who just clicks around in Power BI. Your goal is to help the business make decisions. And decisions are made much more easily when looking at a clear, precisely designed picture, rather than a spreadsheet with thousands of rows of numbers. Our job is information distillation—we process gigantic datasets to ultimately generate that one chart that precisely points a finger: “Here is our problem, and here is where we are losing money.” Business is a journey from one solved problem to the next fire. We are the navigators.
6. Writing and Speaking in Simple Language (Soft Skill)
We associate experts with people who can build multi-layered sentences and juggle difficult terms. In the data world, it is very easy to fall into this trap. An analyst who walks into a meeting and says: “Our LTV to CAC ratio is not optimal enough to escalate this cohort segment due to increasing churn” might feel smart, but no one understands them.
A company is a complex organism full of people with different backgrounds. To be heard, you must speak simply. I am reminded of a brilliant scene from the movie “Margin Call,” where the powerful CEO of a massive investment bank turns to a young analyst: “Speak to me in plain English. Speak to me as you might to a young child or a golden retriever. It wasn’t brains that got me here.”
The crux of the matter is that business problems are terribly complicated. The real art lies in talking about very difficult things in an extremely simple way, but without removing the key nuances. You must be able to capture the pure essence of the phenomenon.
7. Python (Technical Skill)
A few years ago, I would have told you that Python is a nice addition for an analyst, but completely optional. In 2026, the situation has changed. Knowing the basics of Python is now essential for two fundamental reasons.
First, the barrier to entry has dropped drastically. Tools like Jupyter Notebook or Google Colab are commonplace in most companies. Automating routine, tedious processes using Python scripts is incredibly easy and saves hundreds of hours.
Second, Python integrates phenomenally with the new world of artificial intelligence. The ability to query the API of advanced LLM models, loop data, and automatically categorize thousands of records are tasks that can be a nightmare in Excel or SQL, but take just a few lines of code in Python.
You don’t have to be a software developer right away, but you must master the absolute basics of the language (variables, loops, classes), the rules of connecting via API, and the “Holy Trinity” of data analysis in Python, namely the libraries: Pandas, Matplotlib, and Seaborn. Moreover, the path between a regular analyst and a Data Scientist or Data Engineer has shortened considerably. By knowing Python, you open the door to much more advanced and lucrative roles in the market.
8. Independence (Soft Skill)
Although an analyst’s job is often a team game, independence defines your value. It’s not about sitting in a dark basement and talking to no one. It is about the ability to solve problems without someone constantly holding your hand.
Unfortunately or fortunately, our work involves constantly hitting a wall. We encounter gaps in data, unclear business processes, or technologies that simply refuse to cooperate on that particular day. An independent analyst, seeing a recurring problem, does not go crying to the boss asking, “What do I click now?” Instead, they prepare a “Proof of Concept” for a solution that will automate dealing with that problem in the future.
People in companies are terribly busy. Bothering them with every trivial matter is not welcome. Today, thanks to technological support, an analyst can be a kind of “One-Man Army.” On the one hand, you talk to the business; on the other, you model data; and on the third, you implement AI-driven automation.
9. Artificial Intelligence as a Separate Stack (Technical Skill)
AI has ceased to be a novelty and has become a full-fledged element of our technology stack, as important as SQL or Power BI. This skill boils down to a few key aspects.
It starts with the ability to spot safe automation potential—places where deploying AI will save time, and the risk of hallucination or error won’t cost the company millions. Another issue is proficiency in prompt engineering and context management. You cannot be a “fanboy” of just one model. Today you use ChatGPT, tomorrow Claude, and the day after a locally run Open Source model. You must switch smoothly between them depending on your needs.
The highest level of initiation is managing the knowledge space for artificial intelligence. This ranges from formatting data in structured files (Markdown, HTML) so the model can easily “read” them, to advanced work with RAG (Retrieval-Augmented Generation) architecture. RAG allows you to query internal, closed company documents, which significantly enriches text analytics. This is something you simply have to continuously and actively learn.
10. Continuous Learning and Energy Management (Soft Skill)
Finally, the skill that binds it all together. Development in IT never stops. I remember the market migrations from Tableau to Power BI; I remember when Python suddenly started dominating job postings. “Continuous learning” sounds like a corporate cliché, but it has very hard, practical implications.
Learning is the process of changing behavior based on feedback. You must accept the fact that the knowledge you hold today has an expiration date. You cannot learn just on holidays by taking one course a year. You must constantly move along your development path.
Inseparably linked to this is managing your own energy. To be effective in the long run and protect yourself from burnout, you must perfectly balance the time spent absorbing new technologies with time to rest and utilize the knowledge you already have. It is no great feat to pull all-nighters for a month, only to hate the sight of a laptop for the next two months.
The best of us can format our thought process in such a way that learning is a positive reinforcement, not a chore. I really like the distinction between “I got to” (I have to do it) and “I get to” (I have the opportunity to do it, it’s a privilege). When you shift your perspective and start thinking: “I have a great opportunity to solve complex problems using the latest, fascinating technologies,” this work brings unimaginable satisfaction.
Summary
The market places huge demands on us today. The era of analysts who were merely craftsmen pulling data in SQL is coming to an end. We are expected to have excellent technical skills supported by powerful business and communication competencies.
Acquiring all this takes time and guidance. If you feel that you could use some structure, a plan, and the right environment, remember that you can always check out KajoDataSpace, where we build careers together and master this powerful toolset step by step. Think about which of these ten skills gives you the biggest problem today and make it your priority for the coming months.
If you think this knowledge about how the world of analytics is changing could help someone in your professional circle, please share this article on your social media—on LinkedIn, Facebook, or Twitter. The more aware and well-prepared analysts there are on the market, the better for all of us. Talk to you soon!
The article was written by Kajo Rudziński – analytical data architect, recognized expert in data analysis, creator of KajoData and polish community for analysts KajoDataSpace.
That’s all on this topic. Analyze in peace!
Did you like this article 🙂?
Share it on Social Media 📱
>>> You can share it on LinkedIn and show that you learn something new every day.
>>> You can throw it on Facebook – and perhaps help a friend of yours who is looking for this.
>>> And remember to bookmark this page, you never know if it won’t come handy in in the future.
You prefer to watch 📺 – no problem
>>> Subscribe and watch my English channel on YouTube.
Prefer to read in Polish? No problem!
Other interesitng articles:
