5 Requirements for Building a Strong Data Culture

With virtually all enterprises investing more heavily in data analytics, big data, and AI projects these days, company executives overwhelmingly report that they are trying to shift their organizations to a data-driven culture. However, only about a third of executives say they’ve succeeded.

That’s understandable. Cultural shifts are complicated endeavors, and a shift to a truly data-driven culture is as much about people as it is about technology. Let’s take a look at five areas where executives need to focus on laying the right groundwork for change.

1. Empowerment

Executives can get started on the path to a strong data culture by empowering their teams to not only embrace change but to be catalysts for it. Research has repeatedly shown that employee empowerment deepens commitment within the workplace, which is a key driver of innovative thinking.

Empowerment means giving employees the freedom to experiment with new ideas and fail repeatedly — as long as they learn from their mistakes through data and ultimately deliver amazing products and experiences through this iterative process. Think about Chase’s endeavors to conquer the millennial market through new all-mobile products like Finn. Despite being an incumbent brand in the financial space, the company is reimagining how people interact with its services. Such shifts require experimentation, testing, measurement, and iteration, and they mean accepting and learning from failure before you finally get it right.

2. Environment

Right behind empowerment must come environment. You can empower your teams to innovate with data, but if you don’t create an environment that embraces transparency and enables people to discuss and share metrics across the organization on a recurring basis, your siloed efforts will quickly hit a dead end.

Environmental enablement need not be complicated. Focus on channels and updates. In terms of channels, for example, you’d be surprised how effective a Slack channel dedicated to data can be when it comes to encouraging informal conversations and collaboration around data and new ideas. In terms of updates, sharing top metrics during team meetings can provide needed visibility into what’s working and what isn’t. For example, if there was an increase in Customer Satisfaction Score (CSAT) recorded last week, the full team should discuss it during the weekly meeting, look at the underlying data, and endeavor to find a correlation with something that changed that week, be it a new feature that was released or a marketing campaign that went live.

3. Tools

Of course, any good data transformation is going to involve some careful tool consideration and implementation. The key to leveraging tools to build a stronger data culture is to look for ones that can be broadly leveraged by teams across the organization to source their own insights, track product behavior, and drill down deeper into both areas.

Digital behavior analytics tools like Amplitude and Mixpanel can enable teams to see online customer behavior in action and draw out the actions that matter most to their efforts. For example, insurance companies might use behavioral analytics tools to understand friction points in their customer acquisition flows as well as their claims submissions systems. Identifying friction points within each process empowers teams to optimize their conversion rates, or completion rates, and improve the web and mobile journeys of their prospects and customers. Insights from these tools can also feed back into the product development cycle and help teams build substantially better products and features.

4. Process

Of course, the right digital tools won’t get you far without strong processes for tying the right pieces together. When it comes to establishing a data-driven culture that fuels continual product and user experience improvement, it’s essential to create processes that let teams see how product usage and behavioral data tie in with user research, such as surveys and interviews. In other words, when a team uncovers insights within their analytics, they should be able to supplement those insights with user research to gain a more qualitative understanding of the pattern identified. Likewise, qualitative insights from user research or customer feedback can be validated and better understood through quantitative behavior analytics.

For example, a telecom company might conduct surveys at the end of its customer service conversations, whether via phone or conversational AI, to get feedback on the customer experience. Within this feedback, the company might hear comments like, “The customer support section is frustrating. I wasn’t able to find the answer I was looking for and had to call in.”

With such insight, the telecom company can dig deeper using behavioral analytics to identify patterns among users who express this sort of frustration and build analytical cohorts. By identifying likely friction points for these cohorts of users, product and marketing teams can endeavor to make the support section more intuitive, experiment with different copy, and even consider new ways of presenting the information. Product teams should also then track the success of their improvements. Which leads us to…

5. Metrics

As a final step in creating a strong data culture, teams must be aligned around the right metrics to gauge success and opportunity for improvement on an ongoing basis. They should measure their success on particular initiatives as well as overall customer satisfaction scores.

Meanwhile, the broader organization must align along a North Star KPI that takes into account revenue drivers as well as measures around customer value. For example, a travel company might want to create and align around the metric of Weekly Satisfied Travelers (WSTs), which  can represent a calculation of the number of bookings (revenue) in which a traveler also rated the experience positively (value).

Overall, if executives want their data-driven cultural transformations to succeed, they need to take a holistic approach that extends beyond the data tools themselves. These transformations are not simple endeavors. But in today’s age of disruption, they are essential — and well worth the effort.

Originally published in VentureBeat.