Demystifying the Data Architect

An intern's take on who the Data Architect is, and what it takes to be one

As undergraduates from a business-oriented school, ‘Data Architecture’ or ‘Enterprise Architecture’ aren’t terms we commonly come across, especially outside of the computing faculty. I asked a few peers what they thought my Data Architect team at SAP does and some phrases that I got in response were, ‘some backend data system thing’, ‘building structures through coding’ and ‘devising the infrastructure of data to slay the virus’. One peer even said, ‘you play Minecraft’, which might or might not make sense to most.

Interesting responses aside, the point of the question was to show how shrouded the responsibilities of a Data Architect are among university students. However, given that it’s an increasingly demanded role, I hope to entice more students to explore it with this writing.

Introduction

Data Architecture is one of the roles along the Business-Technology line, counterparts being Solutions Architecture, Product Management, Technology Consulting, Presales, and more. I wanted to explore roles that not only leveraged deep technology expertise, but also provided a strong client-facing element. As a result, I, very gratefully, landed in SAP’s Data Architect team that checked both boxes.

After 8 months of being with SAP’s Data Architect team, here are some insights into data architecture, and just who the Data Architect is.

So, what is Data Architecture?

Data Architecture is essentially the framework that standardises the process of data collection, storage, preparation, modelling, sharing and usage within a company. In simpler terms, it dictates the way data flows through different data storage and management solutions.

To give you an example, let’s think of chess. Each piece represents a company, and the rules of each piece represent the company’s data architecture. A simple architecture only allows the company to move like a pawn – one step at a time. While a state-of-the-art architecture would allow the company to move like a queen – everywhere.

Contextualising the example, a company that imports local excel and CSV files into Tableau to conduct analytics will never match the pace and growth of a company that streams real-time data directly from data sources into enterprise-level data orchestration tools.

Think of Data Architecture as a set of rules and policies on managing data to help businesses run better.

Often, data architectures are represented via lines and boxes on a PowerPoint screen, with varying levels of complexity depending on the stage of the engagement and/or the target audience.

I've included an example of a high-level data architecture below, showcasing how data would flow through the SAP Business Technology Platform’s Data & Analytics solutions.

Then, what does a Data Architect do?

Does that mean the Data Architect just draws boxes on PowerPoint, presenting it to customers as a blueprint of sorts?

Well, yes.

But that’s not the whole story. In fact, it’s only a small fraction of the Data Architect’s role.

Fundamentally, the Data Architect is a customer-centric role. For every customer that goes to them, the Architect aims to help them get from their current infrastructure, to an improved one that solves their most pressing business needs, to a future-ready ideal state incorporating all the best practices of a solid data architecture.

For every engagement, a chunk of the Data Architect’s time is spent understanding the customer’s business priorities and their current system architecture. This is followed by a series of internal meetings with experts from all verticals to devise a proposed To-Be architecture. The next chunk of the Data Architect’s time goes into back-and-forth with the customer. Each meeting aims to both refine the solution, and get the customer closer to seeing that the benefits of the proposal outweigh the costs.

While data architecture diagrams are a common sight in the day-to-day of an architect, they are simply a tool to help customers visualise the logical flow of data between applications.

The complexity behind the Data Architect role

As with every role, while the explanation above seems to imply a simple and standard Understand-Propose-Sell-Repeat cycle, there are many hidden complexities that add to the difficulty of the role.

On one hand, the Data Architect assumes the role of a salesman with strong stakeholder management and persuasion skills. While majority of the customers they encounter face standard issues such as data warehousing and pipelining, many of them harbour preconceived notions of their problems and are, at times, adamant about how our solutions can’t possibly solve them. Breaking that "can't do" mindset will not only take time, but skill.

On another hand, the Data Architect also takes on the role of a technical expert, well-versed in data-related technologies and the best practices of great data architecture. In this aspect, the Architect is tasked to flawlessly execute several tasks – Thoroughly understanding a customer's possibly out-of-hand data situation from just a handful of touchpoints, justifying the benefits of one architecture over another, backing up claims with solution demonstrations of SAP tools, working with other teams to curate Proofs of Concept, etc.

Personally, I think one of the most intimidating parts of being a data architect lies in handling customers that are equally well-versed in data and database technologies, especially given the fast-paced nature of the technology landscape where new technologies are constantly emerging.

As such, to excel as a Data Architect would mean to not only master the stakeholder management skills of a salesman, but also possess comprehensive, and progressive technology expertise to effectively tackle the growing difficulties of managing big data.

Summary

With more and more people entering the technology industry, many are also on the look out for roles that are more inclined towards the business aspect rather than the software engineering one. Hence, to those who share a similar sentiment with me - to those who are keen to pursue a role more client-facing than development yet wish to immerse themselves in an environment that trains your technical knowledge - I encourage you to explore Data Architecture!

While my experiences may be SAP-specific or incomplete, I believe the fundamental nature of the role remains the same across the industry, with no two days being the same. Open to discussion and learning if you share a different view on Data Architecture and who the Data Architect is! 😊

#dataarchitecture #dataarchitect #data

References

• Simplilearn | Who is a Data Architect?
• IBM | What is a data architecture?
• BMC Software | Data Architecture Explained