Every week someone asks me some version of this question: "Where do I start?"

There are a thousand roadmaps online. Most of them are lists of tools and courses that will keep you busy for two years without ever getting you a job.

So today I want to give you my version — not what looks good on a curriculum, but what actually matters in 2026. What to learn, what to skip, and what nobody puts on the roadmap but probably should.

My YouTube video this week goes deeper on all of this — link at the end.

Before we proceed - a small ad. Your clicks on the ads help to cover newsletter hosting fees. And it is a way to say Thank you for the newsletter.

Meet AIApply.

Your AI Career Agent for modern job searching.

AIApply automatically:

Stop manually filling out forms.
Stop rewriting the same application over and over.

Whether you’re actively job hunting or casually exploring opportunities, AIApply keeps your applications moving around the clock.

Let AIApply do it.

Start here — and don't skip it

Excel — yes, still

I know. Everyone wants to skip Excel because it feels boring and old (it turned 40 last year). Do not skip it.

Excel is still the most widely used data tool in the world. Your stakeholders use it. Your boss uses it. You will spend more time in Excel than you expect.

It is easy to learn and it is a good foundation to help you learn other tools and programming languages. Also now Excel even has python embedded, so it has some advancement.

Focus on things like lookups, if, sumif, countif, split to column, removing duplicates, pivot tables and basic charting.

SQL

If you learn nothing else, learn SQL. It is the language of data. Every company stores data in databases. Every data analyst needs to query it. It is also the fastest skill to learn relative to how much it opens up.

You do not need to master it before moving on. Learn SELECT, WHERE, GROUP BY, JOIN, and basic aggregations. That covers 80% of what you will actually use at work.

I have several SQL cheatsheets that you can grab here

PowerBI/Tableau

If several years ago knowing a visualisation tool was a nice to have, now it is a must. As a data analyst you shouldn’t be a professional user, unless you want to be a BI (Business Intelligence) Analyst, but you need to know how to use it and how to build a simple report yourself.

Good news, PowerBI is a Microsoft product, so its interface will remind you of Excel (that’s why I suggest to start with Excel).

Choose PowerBI or Tableau, don’t learn both. If you understand how to work with one, you can quickly learn the other tool if needed. A lot of knowledge is transferrable, the rest is googleable :)

What you can skip — or at least delay

Python — not yet

Before you come at me — I love Python. I use it every day. But if you are a complete beginner and your goal is to get your first data analyst job as quickly as possible, Python is not the priority.

SQL gets you in the door. Python makes you dangerous once you are inside.

Learn the basics so you are not lost when someone mentions it. But do not spend six months on Python tutorials before you have landed a job. Get the job first, then deepen your Python skills on real problems.

(Side note: Python for Beginners Part 2 is coming next week in this newsletter — stay tuned.)

What courses do not teach — but employers care about

This is the part of the roadmap nobody puts on the list.

Storytelling with data

You can run the most sophisticated analysis in the world. If you cannot explain what it means to a non-technical person, it has no value.

Every analysis you do should answer three questions: what happened, why it matters, and what we should do about it. Practice writing that up clearly. Practice presenting it out loud. This skill is worth more than any technical tool on the roadmap.

Communication and stakeholder management

Data analysts do not work in isolation. You will get vague requests, changing priorities, and stakeholders who do not know what they want until they see it.

Learning to ask the right clarifying questions, manage expectations, and push back diplomatically on bad requests — this is what makes a good analyst genuinely valuable. It is also almost never taught.

Presenting your work

Most analysts I have seen make the same mistake: they present their process instead of their findings. Nobody cares how you cleaned the data. They care what you found and what it means.

One clear insight with a recommendation beats ten slides of methodology.

The part of the roadmap that did not exist two years ago

AI is not optional anymore

I have said this before and I will keep saying it: AI is now a productivity expectation, not a nice-to-have.

My husband hires for senior finance roles. He asks every candidate what AI tools they use. If the answer is nothing — it is a no. Because his entire team uses AI, and someone who is not using it is simply slower than everyone around them.

The same is happening in data. Companies expect you to do more in less time. AI is how you do that. Use it to write SQL faster, to check your code, to generate synthetic data, to summarise findings, to draft presentations.

You do not need to be an AI expert. You need to be someone who uses it comfortably as part of how you work.

My version of the roadmap

If I were starting from zero today, this is the order I would follow:

Notice that I put "start applying" at step 5, not step 8. Most people wait too long. You learn faster in a job than in a course. Get in the door and figure out the rest from there.

I recorded a full video on this topic this week — walking through the roadmap in detail, what I would skip and why, and what I wish someone had told me when I was starting.

Watch it here → link

And next week — Python for Beginners Part 2. I got so much feedback about part 1 last week, that I have to do part 2.

One more thing

If you are actively job searching, my friend built something genuinely useful. Dataford has company-specific interview guides — real questions, role breakdowns, and culture ratings for Google, Meta, Amazon, OpenAI and hundreds of other companies. Worth bookmarking before your next interview.

Check it out here → https://dataford.io/interview-guides

Keep pushing 💪,

Karina

Just starting with Python?

Master key data analysis tasks like cleaning, filtering, pivot and grouping data using Pandas, and learn how to present your insights visually with Matplotlib with ‘Data Analysis with Python’ masterclass.

More from me: YouTube | TikTok | Instagram | Threads | LinkedIn | Substack