In a world where an emotionally evocative yet singular experience can garner more attention — and clicks — than a broader reality, data is what separates entertainment from the news.
Axios delivers breaking local, national and international news in succinct bites. The stories provide readers with essential information that matters in its proven Smart Brevity formula.
From understanding audience interests and providing targeted content to generating business insights and honing marketing strategies, data is the underlying force in Axios articles, behind the scenes and across teams. The company has even created a system of data liaisons to build bridges between teams.
But above all, data informs. As Axios strives to deliver the news of the day with brevity and visual appeal, objective insights are the determining, legitimizing factor.
Axios’ commitment to quality data extends beyond its journalists, as Senior Director of Data Grism Bolks explains below. Bolks shares how Axios fully leverages data throughout its user experience and journey.
Why does Axios need a data platform? What gave rise to it and how does this impact the business and its consumers?
Our audience is paramount to our success at Axios. To better understand and cater to their needs, we collect and manage content, audience and business operations data. We developed a centralized data platform to manage and analyze this data effectively, supporting informed decision-making and creating AI-powered products.
By leveraging modern data tools and technologies, the platform gives us the power to analyze data to identify patterns and trends that would have been difficult to detect otherwise. This enables teams at Axios to be more effective at their jobs, whether it be through a consistent and accurate picture of business operations, insights to tuning advertising and marketing strategies, identifying new business opportunities or a better understanding of our audience.
A deeper understanding of our audience allows us to deliver a continuously improving user experience. We achieve this through developing content that resonates with them, prioritizing formats that meet their needs, empowering local journalism and intelligently recommending content or other products tailored to their interests. Our data platform is the key enabler of these efforts.
“A deeper understanding of our audience allows us to deliver a continuously improving user experience. Our data platform is the key enabler of these efforts.”
What role did you play in developing and launching the product? What tools or technologies did your team use to build the product and why?
As a member of the early team at Axios, I have had the privilege of overseeing our data teams’ growth and the platform’s development. When assessing tools and technologies, we consider factors such as time, cost, technical proficiency and the ability to facilitate cross-functional collaboration.
We utilize Redshift, a columnar data warehouse that serves as a central repository for data. This tool is purpose-built for speedy data retrieval for analysis and enables our analysis and business users to quickly access and query relevant data.
We employ Apache Airflow as an orchestration layer and DBT for data transformation, enabling us to create data pipelines and ensure high-quality, reliable data for our business needs. This combination allows us to efficiently leverage our warehouse’s computational capabilities while also fostering close collaboration between data scientists, analysts and engineers.
We also utilize Terraform for effective infrastructure management and rely on Python as our preferred language for everything from data engineering to machine learning development due to its extensive range of capabilities.
What obstacles did you encounter along the way? How did you successfully overcome them?
As our company grew, one of the significant obstacles we faced while building our data platform was ensuring that more business users had the ability to leverage data in roles. Although the team created many dashboards to serve specific purposes, we encountered an analytics bottleneck due to the increasing requests to answer nuanced questions.
To address this issue, we migrated to Looker, a business intelligence platform that enables business users to easily explore, analyze and visualize data using a semantic modeling layer.
This helped us continue providing business users with user-friendly dashboards but, more importantly, reduced technical complexity enabling more self-service analytics. The data team was able to focus on developing more profound insights while also meeting the company’s increasing data demands.
As we migrated to Looker, we adopted a deliberate approach to working with each business team individually. This allowed us to provide training, gather feedback and ensured that we met or exceeded the data capabilities provided by our previous tools. This close collaboration helped us stay motivated and aligned as a team throughout the development process.
What teams did you collaborate with in order to get this across the finish line? What strategies did you employ to ensure that cross-functional collaboration went smoothly?
Axios data product managers play a crucial role in facilitating collaboration and ensuring that the team is focused on solving problems that align with the business needs and strategy.
“Axios product managers play a crucial role in facilitating collaboration and ensuring that the team is focused on solving the problems that align with business needs and strategy.”
The development of the data platform was a collaborative effort, and the data teams worked closely with various product and tech teams as well as business teams like growth, audience development, editorial, sales and executive teams. This collaboration was crucial in aligning everything from data sources to user-facing tools and insights with the business priorities. It ensured that the platform could drive product innovation and be an integral part of day-to-day decision-making.
To promote collaboration and reduce siloing on an ongoing basis between Axios teams, we have implemented a system of data liaisons. These liaisons, who are data team members, serve as subject-matter experts and facilitate communication and ease coordinated development between different teams.
When you think of other companies in your industry, how does Axios compare when it comes to how you build and launch new products?
At Axios, our product design and technology teams operate as cross-functional Scrum teams using an agile methodology, and our data teams follow suit. Our centralized data team consists of two sub-teams: one focusing on the data platform and the other building consumer insights. The teams include product managers, data engineers, analysts, scientists and quality engineers. By centralizing our data teams, we ensure that team members have the support of their peers and use standard data systems and tools.
Quality engineers at Axios also have a unique role on the teams, which differs from traditional quality assurance practices in many other companies. Their role involves designing testing plans and de-risking the most critical aspects of our products. They ensure that the entire team is focused on testing efforts that are most important, resulting in increased collaboration and higher-quality products.
