4 strategies for driving analytics culture change

Executives looking to bring about culture change in their organizations face a long road ahead of them. This is because changing the analytics culture is not as simple as mandating the use of analytics by every person in the organization. Changing the analytics culture requires careful planning and delivery of information that makes the individual employees successful.

Culture change is hard

The solution lies in a mix of tooling and analysis and information delivery architecture. Often, culture changing strategies can fall flat because they approach the problem from purely a tooling perspective. Vendors offering such tools paint a rosy picture of how the right tool can change the culture and behavior within an organization. However, the problem often is more complex.

A top down approach that aims to force the adoption of the tool across the organization doesn’t work. Any directive that impacts the day to day workflow for an employee significantly enough that it hurts their productivity and success goals is bound to generate a lot of push back from the employees. An addition to the set of things an employee has to do will compete with their ability to complete the predetermined tasks and this will force them to de-prioritize the usage of the analytical tool.

Driving culture change requires deep introspection of how the enterprise carries out its business functions. The analysis of how and what information is required to not only complete the tasks at hand but carry them out with a higher quality is a strong precursor. This breakdown of what information is required and the architecture of how it needs to be delivered is critical to this culture change. Information, not data delivered to fit into an existing workflow (and not extend or modify it).

Strategies for information delivery driven culture change

Here are 4 strategies for information delivery architecture that can help enterprises to drive an analytics culture change

1. A la carte vs. a seven course dinner

Choice is the basic premise of this strategy. There are two distinct mechanism for analytics consumption is an enterprise. The ‘a la carte’ strategy enables the user to, in a self directed manner, choose the information they want in the form they want. The “seven course dinner” strategy personalizes, contextualizes and packages information for a user given their context, needs and purpose. In this form, the user consuming the information is easily able to search and discover information they need that they require to complete their tasks.

‘A la carte’ requires a tool designed to offer self service for non technical users who are able to create information from scratch or extend an existing analytic and requires supporting capabilities that enable users to find and focus on specific data sets or specific rows and columns with a given data set. Horizontal analytics products such as Tableau, Microstrategy, Qlik etc. enables any data to be quickly explored and analytics and insights generated.

On the other hand, the “seven course dinner” removes any major obstacles to information delivery however imposes a high cost upfront on the organization to map and transform incoming raw data into a form that can fit into what is required for the predetermined format of the information delivery. This upfront cost can often pose a challenge however if mitigated properly, can very easily equip the entire organization with relevant and required information. Purpose driven analytics products such as NewRelic (for application performance monitoring) or Google Analytics (for web analytics) or Salesforce Wave (for Sales performance) are examples of the “seven course dinner””; that is with some upfront cost of mapping raw data into the format expected by the tool, a huge amount of relevant information to a specific problem can be easily generated and delivered. Typically, the “seven course dinner” approach puts an additional load on the developers building data pipelines to reduce the burden on the users who are interested in insights from this data.

2. Analytics vs. information and insights

Simply enabling tools for the enterprise and hoping for adoption is not realistic. This is one of the main reasons why enterprises get saddled with a whole plethora of analytical tools with minimal usage. As new tools come out, each of them promises to solve the adoption problem and enterprises can get stuck in a 2-3 year replacement cycle as they look for a product that gets wide adoption in the enterprises.

From a data and analytics perspective, there are three types of employees. Those who care about analytics and are able to design and self service their information needs. These users are data savvy and understand the intricacies of collecting, merging, preparing and analyzing data. Next, there are users who care about analytics but are not able to self service and depend on someone else to provide the analysis. Lastly, there are users who simply don’t care about analytics. Driving culture change is tricky because if the entire organization does not align and move together, culture clashes are bound to occur and the added friction will slow down or simply stop any proliferation of a data driven culture in the organization.

In this situation, driving culture change requires a lot of supporting structures and processes to be built in the enterprise. Just as culture change takes a lot of time, users’ maturity and comfort with data and analytics also changes over time. The key is to ensure that users are able to get the support and confidence they need at the pace that works for them. This means that data wrangling and preparation needs to be restricted to a smaller set of people and the majority of the organization is able to leverage the work done by these people to further and satisfy their needs for analytics without having to worry about or understand how low level things are done. In addition, as information is generated, it should add to the body of information in the organization so that the next person looking for the same information simply has to search for it.

3. Come and look vs. notify and deliver

Another common problem that prevents analytics adoption and the subsequent culture change is that building and deployment of tools and analytics is often seen as the end. However, any model of operation that requires a user to navigate to a different destination to begin their relevant information seeking process is bound to put an undue burden on the user to remember to, understand how to and eventually use the dashboard and/or tools.

A more scalable approach to driving the culture change is to couple the analytics and dashboards with notifications of interesting events that are delivered to the relevant set of interested users through a mechanism chosen by the user. This ensures that the user does not have to constantly navigate to a central location looking for something that interests them. Instead, they are able to tell the system what is of interest to them and the system notifies them when there is something interesting for the user available in the system. This reduces the cost of using the system imposed on the user in terms of time and attention and drives usage at the right time.

4. Aid & assist vs. sideline and marginalize

Another key hurdle to culture change is the proliferation of analytics in the organization that attempts to sideline and marginalize certain functions and roles in the organization. This is a critical mistake that can derail any attempts to enforce culture change as the function or role that is getting sidelines and marginalized can begin to work against the tool or analytic either passively (by not utilizing it or helping the analytic get better) or actively (by focusing on the problems and issues).

The key strategy to deploy here is the “aid & assist” model. In this model, the enterprise has to focus on creating analytics that do not attempt to sideline and marginalize key functions but rather focus on information and delivery that makes these functions and roles more productive, efficient and less error prone. Increasing productivity and driving costs automatically will empower these functions and roles to focus on decision making and driving change as opposed to spending time trying to either find the information they need and interpreting what they find to make to the right decision. In the aid & assist model, the focus shifts to automation and process improvement driven by analytics that feeds directly into a triage and decision process.

Conclusion

Be it ‘a la carte vs. a second course dinner’ or ‘analytics vs. information and insights’ or ‘come and look vs. notify and deliver’ or ‘aid & assist vs. sideline and marginalize’, choosing the best strategies to implement a culture change in an organization to be more data driven and agile requires constant attention and introspection of what analytics are offered, in what modes are they delivered, through what mechanisms they are used and in which processes they are internalized.

Key metrics still end up being number of users who are using the analytics and how much time they spend interacting with the system across any of its interfaces and features. However, the leading indicator of a culture change is the time it takes to onboard to the analytics system and the word of mouth and network effect generated from a single employee’s adoption of the system.

Driving culture change is not easy however with right strategies and a data driven approach, it can be made more likely to happen.