4. Data Team Structure and Profile

At GetNinjas, we structured the data team into two main groups: Decision Science and Data Science. This division allows us to leverage data for both business decision-making and real-time model-driven decisions.

1. Decision Science: Data-Driven Business Decisions

The Decision Science team is responsible for helping business teams, especially product managers, make data-driven decisions. This group works closely with product, marketing, and other business areas to ensure the insights gathered from data align with business objectives. Their main focus is analyzing user behavior, engagement data, and revenue metrics.

Key attributes of Decision Science team members include:

• Strong business acumen: They need to understand the company’s goals and question existing processes to identify areas for improvement.
• Technical skills: Team members should know tools like SQL, Excel, and Tableau, and have basic programming knowledge for solving complex problems.
• Profile: The ideal candidates are often interns or recent graduates with degrees in STEM fields. These individuals are flexible and have a fresh perspective, which helps us train them in-house according to the company’s unique data processes and tools, such as Snowplow.

2. Data Science: Real-Time System Optimization

The Data Science team focuses on developing systems and algorithms to enhance the user experience in real-time. This team creates models based on historical data and uses these models to make immediate decisions that improve the platform's efficiency.

Key attributes of Data Science team members include:

• Programming and modeling skills: These individuals need strong programming skills and the ability to create models that can be deployed in production environments.
• Business sense: Although technical expertise is crucial, team members must also understand the broader business goals to avoid focusing solely on analysis and models without knowing the impact on the product.
• Profile: Typically, data scientists in this group have advanced degrees (e.g., PhDs) or several years of experience in the field. They are technically proficient and able to deploy models and systems into production seamlessly.

Evolution of the Data Team Structure

Initially, the BI team was centralized, with a functional structure that served multiple departments, including product and marketing. This setup worked during the early stages but lacked the granularity needed as the company grew.

In 2016, GetNinjas transitioned to a more decentralized model by embedding BI analysts into cross-functional squads. These teams became self-sufficient and focused on specific objectives and key results (OKRs). The BI team members became integral parts of the product development process, working directly with product managers to make decisions based on data.

Example of Team Structure

• Acquisition Team: Focuses on bringing new clients to the platform and improving marketing ROI.
• Professional Experience Team: Works on improving the in-app experience for professionals.
• Payments Team: Enhances the checkout and payment process for users.
• Data Science Team: Develops and deploys mathematical models to optimize product performance in real-time.
• Internal Tools Team: Builds tools for internal teams, such as CRM systems.

Each team, except for Internal Tools, has a dedicated BI analyst. These multidisciplinary teams work with a clear focus on OKRs, making it easier to align their goals and track progress. BI analysts in each team also have regular alignment meetings to share insights and ensure that their work complements other teams' efforts.

Key Takeaways for Building a Data Team

1. Two Core Functions: The data team is split into Decision Science (focused on business decisions) and Data Science (focused on developing mathematical models and algorithms).
2. Multidisciplinary Teams: Embedding BI analysts within product teams resulted in better outcomes because they became active participants in decision-making, providing a quantitative perspective that was previously lacking.
3. Data Infrastructure: The team is responsible for maintaining the data infrastructure, ensuring that data pipelines are efficient, performance is optimized, and data from various sources are integrated.
4. Data Modeling: The team takes raw atomic data and transforms it into more meaningful business data models for use by analysts across the organization.
5. Training and Empowerment: Anyone in the company, from customer support to product, can access data if they are willing to learn. The goal is to democratize data access and encourage decision-making based on data.
6. Continuous Hypothesis Generation: The data team plays a critical role in proposing new hypotheses and exploring data to uncover insights that can lead to improved business outcomes.

Impact on Product Teams

The data team’s work has a significant impact across various areas of the business:

• Marketing: Helps optimize budget allocation and track the return on investment (ROI) across different channels.
• Technology: Monitors load times, identifies bugs, and ensures better app and site performance.
• Support: Tracks team efficiency and helps optimize the commission model.

Empowering all teams to make data-driven decisions ensures that those closest to the problem can generate the most relevant hypotheses and solutions.

By embedding data expertise into each product team, GetNinjas has been able to significantly improve decision-making processes and product performance.