A 6-month Roadmap for Learning Data Analysis

Cody West
Cofounder @ The Query

Cody is a data analyst and analytics engineer with 9 years of experience. He currently works in tech on growth and marketing data analytics.

Table of contents

Now that you know you want to become a data analyst, have committed for the next 6 months, and are set up for success, it’s time to begin learning!

Remember, your goal over the next 6 months isn’t “to become a data analyst.”

Instead, it is to create 7 portfolio projects.

That’s a more actionable goal and will lead to the same outcome you want.

You’ll also have 2 certifications by the end of the next 6 months:

  1. The Google Data Analytics Certification
  2. Tableau Certification OR PowerBI Certification (you’ll pick one)

Note: I do NOT recommend doing all the courses in the Google Data Analytics Professional Certificate one after the other. What you learn won’t stick if you learn that way.

But don’t worry, I’ll show you exactly the resources you should use to learn and in the most optimal order.

Follow the “Learning Roadmap” sections and by the end, you’ll have all the portfolio projects you need for your portfolio AND the two certificates above.

Let’s get started!

Step 1: Data 101 and the Analytics Process

As a data analyst, you’ll be working with data.

Obviously.

So first, what is data?

Think of data as just recorded information.

There are 3 broad categories of data:

  • Structured
  • Unstructured
  • Semi-structured

As a beginner data analyst, you only need to concern yourself with structured data (you can Google the other types if you’re interested).

Think of structured data (aka tabular data) as data that’s neatly organized — like in a spreadsheet.

There are rows and columns.

Each column represents a specific attribute, and each row represents an individual entry or record.

Here’s an example of a table that represents structured data about a list of books in a library.

This could be a simple database table used by a library to keep track of its inventory:

Here are the columns:
  • book_id: A unique identifier for each book.
  • title: The title of the book.
  • author: The author of the book.
  • genre: The genre to which the book belongs.
  • publish_year: The year the book was published.
  • available: A flag indicating whether the book is currently available for borrowing.

This tabular data structure allows for easy searching, filtering, and manipulation of the data.

You can quickly find all available fiction books, for example, or all books by a particular author.

As a data analyst, your job is to use structured data to answer questions that help business stakeholders make more informed decisions by reducing uncertainty for them.

For example, here’s what the data analytics process might look like from identifying the business question to presenting results if you worked in marketing analytics (like me):

  • As a Marketing Analyst, your job is to help the Marketing team make better decisions and report on the results of their work.
  • The Marketing team just launched new ads on Google and your job is to measure the performance of the ads.
  • To measure performance, you work with the Marketing team and determine the best way is to measure the Cost Per Acquisition (CPA) for each ad group (group of ads).
  • All the ads data lives in tables in the company's data warehouse, Snowflake.
  • You write some SQL and pull CPAs by ad group.
  • Then you run your SQL code in Tableau and create a Dashboard showing CPA by ad group.
  • You present your findings to the Marketing team and identify 1 ad group that’s hyper-efficient and make the recommendation that they increase the budget of that ad group.
  • Two ad groups are performing poorly so you recommend shifting the budget away from one of them and then turning the worst performing off completely.
  • The Marketing team agrees and thanks you for the insights.
  • You publish the Tableau workbook to Tableau online and set the data to refresh every night so that the Marketing team can look at their performance dashboard each morning and decide what changes to make to their campaign.

Learning Roadmap: Data 101 & Analytics Process

To begin getting your feet wet with data and learning the analysis process, you’re going to complete the first course of the Google Data Analytics Professional Certificate:

1. Complete Foundations: Data, Data, Everywhere

Next, I highly recommend this 3-part series by George Xing on what it means to be data-driven.‍

2. Read part 1, part 2, and part 3.

Lastly, watch this video on the data analytics process:‍

3. Watch it here

Remember, active learning is the best way to learn! As you go through these resources, continue to ask yourself, “Did that make sense?” If something doesn’t make sense, use YouTube, Google, ChatGPT, or your LinkedIn network to make sense of it!

Step 2: Spreadsheets, Foundational Math, and Basic Statistics

Next, you’re going to learn basic spreadsheet skills along with foundational math (arithmetic mostly) and basic statistics.

It’s best to learn basic math and statistics while you’re learning spreadsheets because spreadsheets provide the environment to apply the math and statistics that you’re learning.

Spreadsheets

Spreadsheets are the best data analysis tool ever invented.

They can do it all; clean data, sort and filter data, visualize data, and more!

Building a strong foundation in spreadsheets will make learning more technical tools like SQL and BI tools easier.

You have two spreadsheet options:

  • Excel
  • Google Sheets

The fundamentals of both are essentially the same, so it doesn’t matter which one you learn.

Google Sheets is free and you don’t have to download anything to start working with it so I recommend starting there.

The first thing to understand when it comes to spreadsheets is that they can do a TON.

There are thousands of features for various applications.

You do NOT need to learn all of these.

As a data analyst, there is a core set of spreadsheet skills you need to learn.

Those are:

  • Filtering, sorting, and cleaning data
  • Structuring data for analysis
  • Conditional formatting
  • Functions
90% of the functions you’ll use are these:
  • SUM, COUNT, COUNTA
  • AVERAGE, MIN, MAX
  • IF
  • COUNTIF, COUNTIFS, SUMIF, SUMIFS
  • VLOOKUP, XLOOKUP
  • IFERROR
If you need to do anything outside this core set of skills, you can easily look it up.

Foundational Math & Statistics

Data analysts don’t need complex mathematics, but you should be familiar with:
  • Basic arithmetic (addition, subtraction, multiplication, and division)
  • Descriptive statistics (average, median, etc.)
  • Weighted calculations
  • Ratios

Don’t go buy textbooks or spend time solving math problems by hand.

All this math will come up as you begin solving data analyst problems, so the best approach is “just-in-time” learning.

If you don’t remember the difference between mean and median and are trying to determine which one to use, look it up on YouTube and learn it right before applying it.

In the learning roadmap below, I’ll also link to a math and statistics refresher mini-course that you can take if you haven’t done math in a while.

Learning Roadmap: Spreadsheets, Math, and Statistics

First, complete the second course in your Google certification:

1. Ask Questions to Make Data-Driven Decisions

Next, go through this 2.5-hour course by Free Code Camp on Excel:

2. Microsoft Excel Tutorial for Beginners - Full Course

Now, it’s time to dive into a few guided projects!‍

Complete these 3 Excel projects:

  1. Full project in Excel - Alex The Analyst
  2. Build an interactive Excel dashboard with Mo
  3. Make an Excel Dashboard in 15 minutes
Finally, the hardest, but most important part — it’s time to create 2 Excel-only portfolio projects.
  • Create the first Excel portfolio project using Kaggle data
  • Create a second Excel portfolio project using Kaggle data
Remember, when you don’t know how to do something or don’t know what to do next, use Google, ChatGPT, and the network you’re building on LinkedIn.
Optional

The rest of the resources are optional, but if you want more practice, here are some other great resources to check out.

If you want to go deeper into statistics:

  1. Check out this Free Code Camp course
  2. Hacker Rank’s 10 days of statistics (goes deeper on knowledge than you need to get your first job, but is well done and an interesting course).
If you want more Excel practice, this course is also excellent.

Step 3: Understanding Databases & Learning SQL

Most of the data you’ll use as a data analyst exists in databases (e.g. Microsoft SQL Server, MySQL, Postgres, etc.) or data warehouses (BigQuery, Snowflake, Redshift, etc.).

You can learn the difference here.

Think of both as just places businesses store data — databases are for more general purposes, while data warehouses are used almost solely for analytics.

SQL is how you access, clean, process, and manipulate that data.

SQL is also your first step into coding, but luckily, it’s one of the easiest languages to learn.

There’s a learning curve, don’t get me wrong, but with 90 minutes of practice for 30-60 days, you’ll be writing SQL fluently.

Here are a few key concepts to remember as you approach learning SQL:

1. 95% of the SQL you’re going to write is in the cheatsheet below.

If you want a PDF or Desktop Wallpaper version of the cheatsheet, I send it out every Thursday as part of The Query newsletter in the “Our Content & Resources” section at the end. You can subscribe using this link.

The other 5% of SQL syntax, you can easily look up with Google or ChatGPT.

It’s helpful to remember that there’s not that much syntax you need to memorize when learning SQL because it makes learning feel much less daunting.

2. When you’re learning SQL, make sure you’re spending at least an hour every day practicing.

You want to build up fluency and the only way to do that is through consistency.

Once you’re fluent, writing SQL becomes enjoyable and that only requires a few months of practice.

3. Remember, it’s going to be difficult.

When I first learned SQL, I had imposter syndrome.

I thought coding was for super geniuses and that I wasn’t smart enough to be good at it.

But after a few months, I started writing it more fluently and then after about a year, I started loving it.

Solving analytics problems with SQL is still one of my favorite parts of my day.

Push through the discomfort in the beginning. Trust me, it’s worth it.

Here’s the learning roadmap for understanding databases and learning SQL.

Learning Roadmap

First, you're going to complete the 3rd course in your Google Data Analytics Certification:

In this course, you’ll get a taste of SQL and use BigQuery (both mine and Kyle’s favorite data warehouse to work with).

If you want a more in-depth course on BigQuery, you can check out Kyle’s free BigQuery for Beginners course on YouTube, though this is optional.

Next, you’re going to go straight into the 4th course in your Google Data Analytics Certification:

Now it’s time to apply your skills with 3 guided tutorials:

In the tutorial above, you’ll also get a taste of your first BI tool (Looker Studio which is free). It’s a much simpler BI tool than Tableau or PowerBI, so it’s great for getting your toes wet before the next section.

Next, complete the 5th course in your Google Data Analytics Certification:

Now, go through this video on YouTube by Chandoo and make sure you’re coding along.

Now complete the SQL Murder Mystery game — it’s super fun and will give you hands-on practice writing more SQL.

Finally, it’s time to complete 2 more portfolio projects!

Both projects should use SQL to process, prepare, and clean your data.

Then you should import your clean data into either 1) Excel / Google Sheets or 2) Looker Studio to visualize it and conduct your analysis.

Use Kaggle or Public datasets to obtain your data.

Remember, when you don’t know how to do something or don’t know what to do next, use Google, ChatGPT, and the network you’re building on LinkedIn.

Here are some optional additional resources, but I still recommend going through:

Step 4: Learn One BI Tool

Business intelligence tools (BI tools) are what you use to create dashboards with pretty charts and graphs.

Usually, you’ll write a SQL query to pull a dataset into a BI tool and then use the BI tool to create a dashboard containing tables and charts for stakeholders (i.e. the people in the business that will be using your dashboard to make decisions).

There are more data visualization tools on the market than there are stars in the sky.

Jk.

But there are a LOT of data visualization tools to choose from.

This can be overwhelming when choosing what to learn.

The good news?

You can’t choose wrong because all of these tools are the same with only minor nuances.

If you learn one BI tool, you’ll be able to learn any other tools you come across very quickly.‍

I recommend choosing one of the following two tools:

  • Tableau
  • PowerBI

Why?

Tableau and PowerBI have the highest demand in the job market.

The other BI tool you’ll see in a lot of job postings is Looker (not to be confused with Looker Studio). If you know Tableau or PowerBI, you can get a job where they use Looker.

So how do you decide if you want to pick Tableau or PowerBI?

First, remember, it doesn’t matter.

You can flip a coin and go with whichever the coin lands on.

If you have a Mac computer, your choice will be easy since you need a Windows operating system to use PowerBI. It’s possible to use PowerBI on a Mac but I wouldn’t recommend it.

Regardless, I recommend going with Tableau because it’s the BI tool you learn while getting your Google Data Analytics Certification, so in the interest of time, it’s more efficient to just stick with Tableau.

Learning Roadmap

First, complete the 6th course of the Google Data Analytics Certification:

If you’re learning Tableau and want a little more practice, go through this tutorial.

Then go through these 3 guided projects:

If you’re learning PowerBI, here’s an excellent tutorial to get you started.

Then go through these guided projects:

Finally, it’s time to create 2 portfolio projects using Tableau or PowerBI!

Both projects should use SQL to process, prepare, and clean your data.‍

Then you should import your clean data into Tableau or PowerBI to visualize it and conduct your analysis.‍

Use Kaggle or Public datasets to obtain your data.‍

Remember, when you don’t know how to do something or don’t know what to do next, use Google, ChatGPT, and then the network you’re building on LinkedIn.

Step 5: Get Your Certifications

Now it’s time to get your certifications!

You have 2 courses left to complete before obtaining your Google Data Analytics Professional Certificate:

  1. A course on R
  2. A capstone project

R Programming

There are two programming languages relevant to data analysts other than SQL:
  • R
  • Python.

In the Google course, you’ll learn the very basics of R.

Here’s my advice — get through the course as quickly as you can and don’t spend too much time trying to learn R.

First, you probably won’t be using R at your first job. You’ll be using Excel, SQL, and a BI tool most likely.

Second, Python is much more common than R nowadays. If you’re going to spend time learning a programming language, I recommend Python.

So why even do this course?

  1. R and Python are very similar so in the future, if you decide to learn Python, this experience with R will help.
  2. To get your certification.

Do you have to get your certification?

No, but it can help (see Chapter 1 for more info).

Capstone Project

In the Google Data Analytics Capstone, you’ll be creating a portfolio project.

I’d recommend following the case study path provided by Google.‍

There’s some high-level advice you’ll get during this course on your portfolio, interviewing, etc. but we’ll go much deeper into all that in the next chapter of this guide.

Tableau / PowerBI Certification

If Tableau is your BI tool of choice, you’re going to want to get the Tableau Desktop Specialist certification.

Here’s how to prepare:

If you follow the advice in these two articles, you’ll do wel

PowerBI Certification

If PowerBI is your BI tool of choice, you’re going to want to get the — Microsoft Certified: Power BI Data Analyst Associate.

There’s a self-paced preparation course Microsoft offers for free if you scroll down on this page:

Also, check out this guide to see what Shungu Dhlamini did to pass the exam. There are additional resources you can use in that guide to help prepare if you want more in addition to what Microsoft provides.

Learning Roadmap

First, you’re going to get your Google Data Analytics Professional Certificate.

You’re going to need to complete the following courses in order:

Next, you’re going to pick either Tableau or PowerBI to get certified in.

If you decide on Tableau, get this certification and go through the following guides to prepare:

If you decide on Power Bi, get this certification and go through the following guides to prepare:

Finally, create 1 last portfolio project.

This portfolio project should be your best yet and the one you put a lot of time and effort into.

Ideally, this is the one that you’re most proud of and that is most exciting to you so you can talk about it in interviews.

This project should use SQL to process, prepare, and clean your data.

Then you should import your clean data into Tableau or PowerBI to visualize it and conduct your analysis.

Use Kaggle or Public datasets to obtain your data.

Remember, when you don’t know how to do something or don’t know what to do next, use Google, ChatGPT, and the network you’re building on LinkedIn.

Finally, get feedback on this project after completing it. Send it to friends, post about it on LinkedIn, and ask for feedback on how to improve it.

Once you get that feedback, make a few iterations improving it.

Step 6: Non-Technical Skills Needed in Data Analysis

There are many non-technical skills a Data Analyst needs to be successful.

These are the most important non-technical skills in my eyes, but this list is not exhaustive:

  • Communication skills and storytelling with data
  • Critical thinking and problem-solving abilities
  • Business acumen and understanding of relevant industries
  • Attention to detail
  • Teamwork and collaboration

You can’t learn these skills by studying.

You’ll build these by creating portfolio projects and doing analysis.

I’ll discuss each skill briefly below and then send you some resources you can use to upskill in these areas, but remember, applying these skills is the main way you’ll improve them:

Communication Skills and Storytelling With Data

You need strong communication and storytelling skills as a Data Analyst because you need to be able to articulate your findings to non-technical stakeholders.

Humans take in new information best through stories.‍

So if you can craft a narrative (i.e. story) around your data analysis work, you’ll be able to communicate more effectively with the people you need to convince to take action based on the findings of your analysis.‍

Communication and storytelling skills MUST be developed through practice.‍

Here are some ideas for practicing:

  1. When you finish a portfolio project and are working on a write-up, use Loom to record yourself presenting your findings (here’s an example video). Make sure you prepare bullets on what you’re going to say. Even experienced analysts do this before presenting important findings.
  2. Ask a friend or family member to sit with you while you present your portfolio project. Ask them to stop you when they don’t understand something so you can get a sense of how you’re communicating.
  3. If you’re afraid of public speaking and want to work through that fear, sign up for your local Toastmasters. It is one of the fastest ways to get reps at public speaking so you improve as fast as possible.
Also, storytelling is a skill that can be learned. Here are a few book recommendations on storytelling I’ve enjoyed:
  1. Long Story Short: The Only Storytelling Guide You’ll Ever Need
  2. Unleash the Power of Storytelling: Win Hearts, Change Minds, Get Results

Critical Thinking and Problem Solving

The only way to improve critical thinking and problem-solving in data analysis is to do data analysis.

But there are a few things you can do to speed up your development.‍

First, and in my opinion the most important, is to write while you are performing analysis.‍

Clear writing is clear thinking, so when you’re trying to analyze a series of charts, open up a document on your computer and begin writing stream of consciousness of what you’re thinking.‍

Here’s the only guide you need to improve your writing.‍

As you write you’ll start organizing the writing and clarifying your thoughts on the analysis.‍

Many analysts don’t write which is a huge missed opportunity. The people I know that write a lot are the deepest and clearest thinkers, so I HIGHLY recommend this as something you incorporate into your workflow as a Data Analyst.‍

I usually write when defining the problem I’m looking to solve (before I write any SQL or start building any reporting) and then I write after I’ve created the report when I’m analyzing trends and trying to come up with conclusions from the data.

Business Acumen and Learning Your Company and Industry

You’ll often hear how important “context-specific knowledge” is in data analysis.

Here’s what that term means:

Imagine you’re a Data Analyst that is tasked with analyzing data from your company's Facebook Advertising account.
But you don’t know anything about marketing or advertising, let alone Facebook Ads.

To be able to effectively perform this analysis, you’ll need to develop context around how Facebook Ads work.‍

You can learn this as you go and sometimes this is the only option you have.‍

In this situation, it’s important to work with the person at your company who runs the Facebook Ads and to ask a LOT of questions.‍

But if you’re going to be a Marketing Data Analyst, you’ll want to develop marketing context so you’re a more effective analyst.

The same thing is true when it comes to business in general (i.e. the company you work at, and the industry you’re in).
There’s context related to all of that.

And the more context you have, the better analyst you’ll be.

How do you develop context?

By asking questions.

Don’t be afraid to look dumb asking questions.

Do you know who would ask a lot of seemingly dumb questions? Richard Feynman — a Nobel prize-winning physicist renowned for his ability to think from first principles.

There’s even a learning technique named after him called the Feynman Technique.

Richard Feynman would ask a LOT of questions in the beginning when he was learning something to build context and create a foundation for understanding more complex things.

When you get a new job or start working with a new stakeholder, make sure you’re uncomfortable with the amount of questions you’re asking.

That’s how you’ll know you’re asking enough.

PowerBI Certification

If PowerBI is your BI tool of choice, you’re going to want to get the — Microsoft Certified: Power BI Data Analyst Associate.

There’s a self-paced preparation course Microsoft offers for free if you scroll down on this page:

Attention to Detail

When you’re working with data you need strong attention to detail.

If you don’t, you can report on false numbers which can contribute to stakeholders making bad decisions.

But there’s a fine line between attention to detail and perfectionism.

As an analyst, I always double and triple-check numbers before reporting on them.

But I don’t double and triple-check while I’m building so I don’t interrupt my flow.

And I don’t check more than 2 or 3 times — that’d be a waste of time.

Another tip I have is to document what you do as you’re doing it.

When you’re conducting a complex analysis over a few weeks, it’s hard to remember what you did 2 weeks prior.

If you document the actions you took, what you learned, etc., that will improve your attention to detail (because you’re writing) and will save you time when you inevitably have to look back on some SQL you wrote a month ago.

Lastly, remember the old military saying, “Slow is smooth, smooth is fast.” That’s the case when it comes to good attention to detail when doing analytics work.

You don’t want to be erratic in your approach.

You want to be slow and methodical because, while it may feel slow at the moment, it’ll end up being the fastest approach.

You’ll avoid making mistakes or confusing yourself by trying to work too fast.

Teamwork and Collaboration

As an analyst, you’ll work with many different people from different parts of your organization who have many different personalities.

To learn how to work effectively with people, there’s only one book you need to read:

There’s a lot more to working with people in the context of a Data Analyst, but you don’t need to learn any of that now.

The book should be enough.

That's it for Chapter 3!

In the next chapter, you'll learn how to land your first job as a Data Analyst.

Click the button below to keep reading :)

Next Chapter:

How to Land Your First Job as a Data Analyst

In the next chapter, you’ll learn how to learn.
Cody West
Cofounder @ The Query

Cody has worked in data for 9 years as a data analyst and analytics engineer focusing mainly on growth and marketing analytics.

He has a wide breadth of experience ranging from working on a 40 person analytics team to being the first analyst at a startup to build the analytics function.

He’s also a data entrepreneur and has built and sold 2 companies in the marketing data space. Currently, he works in tech on growth analytics and GTM strategy.

Cody lives in a log cabin in the woods an hour from the nearest grocery store thanks to Space X’s Starlink internet.