Yup, that’s me. The guy who failed the dbt Analytics Engineering Certification.
You’re probably wondering how I got here.
Honestly, it was mostly because I wasn’t prepared. Had I read a blog like this, I would have had no problem, which is precisely why I wanted to write this blog.
The good news is that this story has a happy ending (I passed the 2nd time with flying colors). And even more good news, I’m going to share all my learnings so that when you take this test, you’ll have everything you need to pass the first time around.
In this blog, I will provide you with the answers to set yourself up for success when taking the dbt Analytics Engineer Certification.
Overall, the difficulty of this test isn’t too hard. I would say the test is a three out of five on the difficulty scale. The test covers a fair range of activities that you perform within dbt. Some of the activities may be tasks you have completed a handful of times, like building out and automating your pipeline. Other questions will cover frequent activities like model compilation or documentation.
The challenging part of this test is the unique question techniques that the dbt team uses. The questions that are most talked about are their discrete multiple-choice questions. These questions will ask you a question, and then one by one, the test will present you with the options and you have to answer for each option whether it is the correct or incorrect answer.
These questions can be difficult, and even cause you to overthink the right answer.
Next will be the hotspot questions, which will present you with an image (either a code or properties file). You will be prompted to select the location of an object or identify an error in the image.
I would say these questions are easier than the discrete multiple-choice questions, and you can move your choice if you realize you picked the wrong area of the image. Aside from the hotspot and discrete multiple choice, the rest of the questions are very standard such as multiple choice, fill-in-the-blank, matching, etc.
What to Study
The dbt Analytics Engineering Certification covers a wide array of topics including:
- dbt model development
- dbt model debugging
- Monitoring dbt pipelines
- Building dbt tests
- Deploying dbt pipelines
- Creating/maintaining dbt documentation,
- Promoting code with version control,
- Establishing a dbt environment
With this plethora of topics, it is easy to find areas that you should brush up on and be prepared for. Thankfully, dbt has a learning path to help you prepare for the exam.
How to Study
The best starting point for review, is to work through the learning path laid out by dbt within their study guide found on the certification landing page. Even if you’ve worked with dbt for some time, you will find some of the content and recommended reading enlightening, which will help you improve your foundation even more.
Aside from the reading, the learning path lays out a path that should be used to work through the free courses that dbt provides.
From my experience with the test, taking all of the courses and reading all of the content will help you have a solid foundation for passing the dbt Analytics Engineering Certification.
Working with dbt, and understanding the nuanced differences between certain types of properties for documentation or specific dbt commands can make the difference between passing and failing the test.
How to Practice
Once you’ve prepared for a couple of unique question methods, you’ve worked through the learning path, and strengthened your foundation, it’s time to begin practicing!
I’m personally a fan of reinforcing the learning with some practice. As an example, you can go find some open-source data sets, load them into your data warehouse, and begin building out a transformation pipeline.
My reason for recommending this is that the test will cover some activities that might be rather set it and forget it, or you’ve not had to debug often. So by practicing, especially in these areas, you can make sure you know not just the correct answers but how to debug any potential problems that might be presented to you during the test.
Some examples of activities you might want to perform:
- Building out source files
- Building out properties files
- Overloading schema/database name generation
- Creating a custom materialization
- Building out a pipeline
- Add SlimCI to your pipeline
In summary, this is what my experience with the exam was like and how I changed my approach to pass the second time.
Overall, this test isn’t extremely difficult but it can be tricky due to differences in testing methods that are used in this test. As always, make use of all your resources from dbt, as well as the phData resources that can help prepare you for the certification.
Need help getting started with the Snowflake Data Cloud or dbt? Our team of data experts are happy to assist. Reach out today!