Understanding the Business Model

The first step in my onboarding process involved gaining a solid grasp of the company’s business model, key revenue streams, and the performance indicators used to assess success. This foundational knowledge proved invaluable during discussions about A/B testing and helped me select relevant KPIs to monitor. Moreover, it enabled me to prioritize tasks based on their potential impact on revenue-related KPIs and to articulate to stakeholders why certain requests needed to be postponed.

Learning the Data Landscape

Next, I focused on familiarizing myself with the data available in our database. I examined tables, their field values, and the ETL processes to understand how tables interconnect and the business logic behind KPI calculations. Consulting data dictionaries and lookup tables was critical in deciphering company-specific field values, such as product identifiers and pricing codes.

Grasping the data early on offered several advantages when engaging with stakeholders. I could immediately determine the feasibility of requests based on data availability, rather than needing time to research afterward. Additionally, I was able to propose alternatives using different data sources, allowing for quicker resolution of requirements in the same meeting.

Networking with Colleagues

It’s beneficial to schedule brief 30-minute introductory meetings with colleagues you are likely to collaborate with. If you’re uncertain about whom to reach out to, ask your manager or stakeholders for recommendations. These meetings serve as an excellent opportunity to break the ice and clarify whom to consult about specific issues. With many employees working remotely, this step is particularly advantageous for new data analysts during their onboarding.

Challenging the Status Quo

As a newcomer, you possess a unique perspective that can reveal problems that long-term employees may overlook. Just because a company has traditionally operated in a certain way doesn't mean improvements aren't possible. Inquire about pain points your stakeholders or colleagues face and identify opportunities for enhancement.

For example, when I began as a data analyst at my current organization, I discovered that the ETL process updating the subscriptions dashboard was divided into four separate jobs. This inefficiency required manual intervention if any job failed, leading to delays in reporting to stakeholders. I streamlined the ETL into a single job, significantly reducing the runtime and simplifying the process — I only had to rerun one job instead of four in case of failure.