A recurring challenge that data teams express in my conversations with them is the persistent struggle with data literacy across their organizations.
For those who may be unfamiliar,
Data literacy refers to the ability to comprehend, interpret, and effectively communicate data insights to others — think of it as fluency in the language of data.
Regardless of your data team's size, we've compiled a list of strategies to help you elevate data literacy and empower individuals throughout your company.
Let's dive in! 🤿
A data warehouse is an indispensable tool for any data team, no matter the size.
Even for a single data analyst, it quickly becomes a game-changer, centralizing your data and enabling seamless connections between various datasets. You can start with a copy of your production database (perhaps in Postgresql), but you can also consider adopting a dedicated analytics solution, such as BigQuery, Snowflake, or Clickhouse.
Side note: for small data teams BigQuery can be a good option with the large free tier AND a smooth integration between GSheets and BigQuery.
Providing access to reliable data is crucial for fostering data literacy within your organization.
A selection of well-crafted dashboards featuring key team and organizational KPIs encourages business users to incorporate data-driven insights into their daily decision-making.
However, avoid the common pitfall of creating an overwhelming number of dashboards in response to every new inquiry. This will lead to confusion, difficulties in locating the right dashboard, and challenges in maintenance for your data team.
Focus on maintaining a few core dashboards and conduct ad-hoc analyses for non-recurring questions to preserve trust in your data.
Ad-hoc analyses serve as a powerful demonstration of the capabilities of your data team. Business users may not always be aware of the potential of data-driven insights, but in-depth analyses from data experts can break down barriers and enhance data literacy. This is the time where you can show best practices and do things only your team can do.
While the prospect of conducting ad-hoc analyses may seem daunting and time-consuming at first, rest assured that it becomes more efficient with practice. Maximize your efforts by repurposing previous analyses and updating them with current, accurate insights. This approach will not only save time but also ensure the delivery of relevant, high-quality information.
Having provided high-quality dashboards and impactful ad-hoc analyses, you'll find that some users need more flexibility than dashboards can offer but may lack SQL proficiency. In all companies, you'll find groups of people using spreadsheets here and there. Rather than fighting this behavior you better support their needs and make it easy to extract prepared and up-to-date data from your warehouse. This approach not only fosters goodwill but over time it will help you identify frequent data use cases for future integration in the transformation step or in dashboards.
The debate between asynchronous and synchronous work has been ongoing, but a combination of both often yields the best results. Office hours are an excellent way to blend these approaches and boost the effectiveness of asynchronous work between business and data team.
Data teams can set up dedicated time slots during which business users can ask questions about dashboards or specific data concerns.
These sessions not only facilitate knowledge sharing and demonstrate how data tasks are executed but also help forge connections and potentially identify future "Data Champions".
Many teams have successfully implemented this strategy, as illustrated in Gojob's webinar on their "Open Data Sessions".
In conclusion, cultivating data literacy within an organization is a gradual process. The strategies outlined in this newsletter can significantly accelerate and streamline your efforts. Always remember to strike a balance between data preparation, facilitating self-serve solutions (including dashboards and, yes, even spreadsheets), and conducting data analyses yourself. These crucial components should be consistently integrated into the workflow of any data team, irrespective of size.
Do you concur with these recommendations? or do you have alternative insights? I am eager to hear your thoughts and suggestions! Ping me to continue the conversation