10 principles of data-driven companies

Data and analytics are essential to objective, informed decision making. With the right practices in place, companies are able to harness the power of data to provide better service to their customers, optimize their supply chains and understand the effectiveness of their marketing efforts. 

Though the benefits of using the right data in just the right ways are obvious, many companies don't have the right principles in place. The result? Up to 70% of projects don't come to fruition.

Use these 10 guiding principles to optimize your org to be truly data driven.

Often, organizations will collect a ton of data without considering why—resulting in too much information to make sense of. Spend time at the outset figuring out what answers you’re looking for, why, and what you’ll do with those answers when you get them. 

Next, figure out where the answers will come from. The monthly dashboard or spreadsheet you use to find answers is likely riddled with underlying manipulation and transformation, as data analysts manually create spreadsheets every month using data queried from a database where they’ve applied several layers of business logic, and that has been enriched with third-party information.

When gaps or quality issues are uncovered, data-driven organizations take the time to rationalize the problem. This can mean going back to your systems and enforcing more rigid requirements on data entry, or building a new system to capture the desired data. It can also mean more clearly defining the transformation and rationalization steps that need to happen with data before it becomes intelligence.

Once the components are in place to understand what to measure and why, where the information comes from, how its captured and what you need to do to it, it’s time for the who. Modern analytics strategies in particular tend to originate from the business side, meaning that IT is often forced to manage a solution they had no say in. A lack of roles and ownership results in scenarios where projects never get sign off and the infrastructure doesn’t get managed. 

Once you’ve figured out the who, you're primed to execute on your data initiatives.

This is the fun part! Choosing the right display to present and explain data is critical to ensuring that the data gathered is met with understanding and utility. Start with an understanding of which chart best represents how the data should be interpreted. If you’re trying to measure a single measure comparatively, for example, consider a bar chart. A scatterplot is perfect for multiple measures and a line chart works great for time-based analysis. 

A single dashboard can have more impact by tending to core design elements like layout, color, typography and size, and is how data visualization goes from “good” to “great”. Some tips:

    • Don’t overuse color. It should be used purposefully with mindfulness about the absence of            color to ensure the data gets seen.

    • Font selection and what words get chosen as companions to the dashboard should have                equal importance.

    • Keeping size in mind ensures that content gets used.

    • Avoid design that is over-the-top and chaotic.

Stories are memorable. They inspire. They communicate the journey and reason behind analysis, and help create empathy and attachment to the outcome. Sharing data stories can become a natural template to inspire new analytics projects, and enhance overall communication and collaboration between different workgroups and silos. 

One great way to get started sharing stories is to start a monthly analytics meeting. You could also send out newsletters or hijack a town hall to give a quick update on your analytics projects.

Analytics done right leaves the audience wanting more. “More” can mean a lot of things—and it’s all about future initiatives or enhancements to analytics projects. Focus on these three keys:

1. Delivering more insights

    Good analytics starts by answering questions and begs for more insights to be found: more     precise questions, more granular details of a subject area and gathering insights for new     subject areas.

2. Deeper questions

    Instead of monitoring and measuring segregated subject areas, you should start to find     relationships among them. Deeper questions are also those that veer into the world of data     science, where the conversation shifts from “What happened?” to “What IF this happens?”

3. Enriched data

    This means trying to fill in those data gaps identified earlier.

It can be easy to get overwhelmed by the notion that everything needs to be built and perfected on the first go. Data-driven organizations know that data analytics is iterative. That means it’s a process that can and should be repeated. Focus on starting with and connecting the business understanding, layering in the data understanding, taking time to prepare the data and associated analysis, then evaluating the results and deploying. Along this path, it’s not uncommon to go back, start over or repeat the cycle a few times before something is finalized.

Measurement is the key to understanding user behaviors, from consumption to creation. It’s also key to being able to plan for the delivery of the “more”, and is your first line of defense for knowing how you’ll need to scale. The three recommended subject areas to focus on are: user engagement, utilization and performance. 

Champions are known for their leadership and evangelism, which means they’re able to amplify the voice of analytics. They understand the mission and the goal—using data to drive decisions. They know that weaving data and facts into strategy is the competitive edge your organization needs to thrive. They’re key to promoting utilization, adoption and awareness around analytics initiatives. 

Once you’ve identified potential champions, continuously work to get them more connected to analytics projects and more involved in the strategy.

Data and analytics are a journey! You can learn more about how to make it a smooth one in our expert-led, on-demand webinar: Four steps to building a data-driven culture.