AWS Data Engineer Associate and Data Analytics Specialty Certifications Compared

Intro and Context

Let me start by saying that I usually wouldn’t compare two AWS certifications beyond considering common services (e.g. how much of my exam prep for X will have helped for certification Y) as I think AWS do a good in logically separating the certifications as long as you understand the differences in practitioner, associate, professional, and specialty levels. I also typically take each learning experience independently and try to cover all material even if it’s part of a previous certification.

However, alongside the announcement of the new Certified Data Engineer Associate (DEA-C01) certification in November 2023, AWS announced they will be retiring AWS Certified Data Analytics - Specialty (DAS-C01) in April 2024, stating “This retirement is part of a broader effort to focus on certifications that align more closely to key cloud job roles, such as data engineer.”

Having undertaken the Certified Data Engineer Associate beta exam in December, I thought it would be worth sharing my views on comparing the two certifications.

Exam Guide Comparison

Reading through the exam guide for DEA-C01, it felt familiar, and I don’t think this is just due to domain experience. The reason it felt familiar to the Data Analytics Specialty (DAS-C01) content was evident when looking at the list of services, but even at a higher level of certification domain you can see there would be a large degree of overlap.

In terms of services in the exam guides, the notable differences are the additions of AppFlow & Managed Workflow for Apache Airflow (MWAA), Cloud Financial Management (Budgets, Cost Explorer), AWS Batch & Serverless Application Model, Amazon Keyspaces (for Apache Cassandra), and AWS Developer tools (Cloud9, CDK, Code*) for the associate data engineer certification.

The overlap is unsurprising given the link between data analytics and engineering, as well as the fact that the data analytics specialty focuses on a lot more than just analytics. That said, the real reason I wanted to call this out is that the data analytics specialty certification recommends 5 years domain experience (compared to 2-3 in DE and 1-2 in AWS for data engineer associate).

It’s worth mentioning that there is a distinct difference between associate level and specialty level certifications. As mentioned above, that starts with suggested experience, which I think this Adrian Cantrill page covers quite well, but also in the exam format. The data engineer exam guide indicates 65 questions (50 scored) and although the exam duration isn’t specified, it’s safe to assume it will be 130 minutes as is the case for the other associate exams, compared to the same question structure but a 180 minute duration. As you might expect, that typically means lower complexity questions for the associate level exam.

Exam Experience

I appreciate this is only useful to anyone who has previously undertaken DAS-C01, or already prepared for it, and is considering DEA-C01 in future. If that’s not you, just skip this section.

Though the maximum durations above indicate less time needed for the data engineer associate certification, I don’t think it’s completely representative of the jump in difficulty. I found time not to be tight for associate level, finishing each with a good amount of time to spare, but only finished the data analytics specialty exam with a few minutes to share. Difficulty aside, I found the exam experience a lot more taxing and needing a lot of mental stamina

In general, the certified data engineer associate questions are centred on less complex solutions (i.e. fewer interacting components or requirements to assess)

In line with the FAQs around why AWS are retiring the data analytics specialty certification, it felt like questions around overarching solution design were much less common in the data engineer associate exam and replaced with more developer type questions (e.g. what would you do as a DE, how would you configure something or code you would write)

Data engineer associate questions follow a more typical format (e.g. least operational overhead, lowest cost). Though these terms also appear in the specialty certification, the ties from question to correct answer aren’t always as clear in the sense that things are less obvious (e.g. could me multiple correct answers, but a specific requirement for queue ordering, throughput, or access controls is slightly different)

There were specific topics with different levels of focus. Compared to the data analytics specialty, my beta exam for certified data engineer associate had

• A similar level of focus on data types, formats, and storage
• Less focus on DB scaling and deployment mechanisms
• Slightly less detail on Streaming solutions - still a key topic, but less depth required. Specifically, nothing on MSK or KCL, and very little on error handling
• Similar areas covered in security and governance such as row level security, IAM, cross account access, but less around encryption
• Much less (very little) on EMR and open-source spark applications
• More focus on Glue and Lambda
• Additional questions covering AppFlow, MWAA, AWS Batch, Cloud9, and Cost Explorer
• Similar coverage on database services
• Much more focus on handling PII
• Some specific questions around SQL, Regex, and hashing

This list is not exhaustive, just the areas I found more notable.

My Thought on The Change

In all honesty, I don’t see this as particularly positive or negative. I personally found the experience of achieving the data analytics specialty certification to be rewarding, but I often comment on the fact that it covers a large range of topics within the data domain, not just analytics, and although it requires lot of solution architecture knowledge that not all data practitioners may have, or want to have, it will add value in some way to most people working in the data domain. Naturally, the question that jumps to mind is; what certifications should data practitioners take if they aren’t engineers or scientists?

However, role-specific / aligned certifications are certainly clearer in terms of deciding the appropriate learning and certification pathway to follow and recognising the relevance related to your day job, and I think there good options, both AWS and non-AWS, for those working in data that may not be well aligned to the AWS developer, data engineer, or ML specialty certifications. Also, the collateral and training for the data analytics specialty certification isn’t going anywhere, and will still be valuable for those who need that level of depth, which isn’t covered as part of the new data engineer associate certification.

With all that said, AWS’ intention of aligning a certification to the role of a data engineer is clear and easy to understand. I’m very interested to see what this means for potential changes or additions to other associate and specialty level AWS certifications as well as how the data engineer associate exam matures over time. If my beta exam is anything to go by, AWS certified data engineer associate will be a worthwhile certification for existing or aspiring AWS Data Engineers.