โœ•
_ Blog / Management

The Ultimate Guide to Building Successful Analytics Engineering Teams

by Dorottya Csikai
/ February 1, 2023
#Management
Gopal Erinjippurath

The Ultimate Guide to Building Successful Analytics Engineering Teams

Building a successful analytics engineering team is challenging. 

First, you have to determine the team’s unique set of success metrics. Then you have to find and hire the right people. Finally, you need to find the tools and processes that fit your team.

How can you approach each step?

Gopal Erinjippurath, CTO and Head of Product at Sust Global, shares valuable insights on how to build and manage analytics engineering teams. He talks about the versatile aspects of analytics engineering, the characteristics he’s looking for when hiring, and the everyday processes and common pitfalls of such teams.

This blog post was written based on episode 77 of the Level-up Engineering podcast hosted by Karolina Toth

This post covers:

What do analytics engineers do?

Multiple functions

Analytics engineers belong to a broad category of individuals who transform data-related inputs into actionable, meaningful analytical outputs to users. 

Finding meaningful outcomes from working with large-scale data requires a few components in place. Therefore, analytics engineering is a combination of essential business intelligence, data science and data engineering. 

Analytics engineers also set up services so that people can access information about predictions or past events. This information should be easy for different groups of people to use, such as analysts, engineers, or executives. This is usually done by creating services or small services that involve both development and operations (DevOps) as well as work on the backend and platform. 

What are the success metrics for analytics engineering teams?

Determine your purpose first

At Sust Global, our purpose is to enable every business decision to be climate-informed over the next 10 years. To achieve this, we need to seed essential climate data analytics and infrastructure into existing and new workflows. The primary outcome we strive for is to effectively use data, machine learning, and software-driven cloud native services to get businesses to use climate data and make climate-informed decisions. 

Different companies, different needs

It’s challenging to determine and track success metrics, especially across companies that are at varying stages of the life cycle of becoming climate-informed and using climate analytics. Some companies might be thinking about a simple demo, some might be thinking about proof of concept, some might be thinking about a scaled operation, and some might be thinking about thousands of operations running in the cloud. 

To track this, we decided to look at what people are actually doing and what's a common thing all these customers care about. We found that all of them seek understanding. 

In our case, one of our flagship products is physical climate risk analytics, which we run at the property level. We are pioneers in using machine learning and data-driven techniques to transform frontier climate modeling into actionable hazard level insights at the property resolution. When our customers are doing these assessments at the property level, they're doing that at varying levels of scale. If we were to accumulate the total number of assets that are getting processed over time, we have a benchmark of how well we are growing our footprint in the market and enabling our customer base. 

Customers and assets

The primary metric that we use to track our success is the total accumulating number of assets processed over time, which should be trending upwards and have an accelerating growth rate. The second metric is the number of new customers who are creating new assessments through our product, either through the API or through our dashboarding product. 

A combination of these two metrics allows us as a software engineering team to track how effective we are in enabling this broader transformation in the business landscape from a lack of climate awareness to climate understanding and then climate-driven action.

What type of engineers do you look for when you build an analytics engineering team?

Knowledge & passion

We’re a mission-driven team that wants to serve the environment and businesses as well. We’re looking at business impact and expertise, so we like to hire engineers who have the required skill set of climate modeling, remote sensing, machine learning and platform engineering and who also resonate with our mission. 

There’s an increased drive across junior and mid-level engineers to work in a climate-related field and to use technology to combat the climate crisis. There has never been a better time to get into the field of climate, and it benefits both parties. The companies are very lucky to have access to this kind of talent, and engineers who want to work in this field can choose from a lot of jobs recently. 

Hiring

People have different passions, and it’s totally understandable if someone is more interested in other fields. Working with climate-related data is more about curiosity and being open to learn the depths of this industry. You can easily spot this curiosity by the type of questions candidates ask from you. 

Motivation, on the other hand, is key in this role. You could be the most skilled, most competent platform engineer or data scientist, but if your motivations aren’t aligned with the team, it hinders the whole team’s success.

What are the processes of your analytics team?

Our processes are somewhat standard. We do biweekly sprints and scrum ceremonies like most companies. Since we work remotely, we've adapted these processes to make sure we have the right amount of meetings, but not too many. This way we have enough time for effective work and collaboration as well. 

We make sure everyone knows what others are working on and where they need help. We also have dedicated time for deep dives into specific topics. Every week we have two hours where we discuss problems and interesting topics that someone on the team wants input on.

We try to create a culture where people can collaborate and share ideas. This has been really helpful because sometimes the best ideas come from unexpected discussions and exploration. It allows us to make new discoveries and learn new things. 

Climate analytics requires bringing together different types of expertise. There isn't just one person who is an expert on climate, there are experts on different parts of it, like forests, oceans, and the atmosphere. But when we think about climate, it's all connected and we need a team that can work together to understand the complexity.

Cross-functional collaboration


We want our product and engineering teams to work closely together. Our goal is for every engineer to think like a product person first, and then be able to turn their code into potential business outcomes that can make revenue for the company. 

We have a growth team led by my co-founder and CEO, whose goal is to turn our innovations into successful business results. This helps us bring new products to the market and grow the company. 

We keep our structure simple by dividing aspects into product, tech, commerce, and growth. This worked well when we had 50 people. However, as we scale up, we need to bring in more expertise into areas like operations, customer success, and human resources.

What tools do you use to support the analytics team?

We use Google Cloud services to make our work easier. We also have tools like Segment to track events on our platform and Mixpanel to analyze them. 

For our software, we prefer standard tools like GitHub. We use Jira to manage our projects and make sure everyone is on the same page. To keep in touch with each other, we use Slack.

All of these tools help us work together even when we're not in the same place. This has always been an important aspect to us because we started our business during the pandemic and needed to work remotely. 

What are the common pitfalls of working with an analytics team?

Clarity

Having clarity on business outcomes is a common pain point in analytics engineering. Engineers often build things based on their own ideas, but their perspective may not align with the needs of customers or the sales and product teams. 

To ensure that the team is building the right things, it's important to provide them with as much information as possible about the business and the customers. This helps them to understand the constraints and guidelines for the project, and to use their creativity in the most effective way possible. 

It's also important to share this information with the team as early and often as possible, so that everyone is on the same page and working towards the same goals. This ultimately helps the team to make the best decisions for the platform, software or analytics survey.

About Gopal Erinjippurath

Gopal is the CTO and Head of Product at Sust Global. He started out as an electrical engineer, and now works in geo data science.

He has worked in different management roles for teams of different sizes for the past 10 years. He’s passionate about software engineering and is excited to share what he has learned throughout his career.

What You Should Do Now

๐Ÿ‘‰ If you are serious about becoming a great engineering leader, you should take a deep dive into the State of Engineering Management 2022 report.

๐Ÿš€ Need developers for your team or project? Hire our experienced Angular, React or Node.js developers! Click here for a FREE consultation.

About the author:

Dorottya Csikai is a content marketer at Coding Sans. She has experience in journalism, interviews, management and content marketing. She researches, writes and edits posts for the Engineering Leader's blog.