ten Principles for getting value from your analytics

How many times have you thought, or heard from your colleagues, that your organization doesn’t seem to be getting the value out of analytics to warrant the investment you are putting into it? Maybe the first step to solving this problem, and supercharging the delivery of value from your data, isn’t about build-versus-buy decisions, architecture tradeoffs, or governance. Maybe we need to take a moment to think about what analytic value is and how it shows up in our organizations in the first place. Here are ten principles to guide our organizations so that we get the most out of data and analytics.

·    To compete on analytics, you should WIN with analytics around your differentiator. Analytics is a broad discipline, and it is impractical to try to be the best at all of it. The place to focus your analytic attention is your business differentiators, whether they be customer experience, operational excellence, or product quality. This is the lens by which you should judge placing analytic resources and infrastructure improvements. Are you needing a robust BI layer that can perform event analysis around customer journeys; visible, self-service dashboards for operational monitoring; or nimble feedback loops via the Internet-of-Things? Demand will always outstrip supply when it comes to analytic capacity because of the simple truth that you will always have more questions than answers – if you didn’t, you’d have a fully automated enterprise. So, to best deploy your investments, ask big questions and streamline data pipes that support the areas of your business that mean the difference between winning and losing.

·    Analytics maturation is a journey, not a destination. Being data-driven in customer interactions and internal operations is a discipline and a practice that never ends. As market demands evolve, so do tools evolve; and as tools evolve, the processes that use them must evolve. Therefore, capabilities and behaviors must also evolve along-side these changes.

·    An organization is only data-driven and analytically savvy when its people are data-driven and analytically savvy. Given that analytics maturation is a discipline rather than a end-point, there is no magic wand to wave that makes a data savvy organization. While there will be analytic experts in the organization, participation in the analytics process will be part of everyone’s role, to various degrees. The more people involved in and using analytics, the more the habits that make up your culture become data-infused.

·    Data is and should be viewed as an asset. Traditionally - unless they sell data or gain revenue primarily from advertising - many organizations have viewed data use and management as a cost: the costs of infrastructure, maintenance demands, and the salaries of the people who wrangle and use the data. Data moves from “liability” to “asset” when it is used. If everyone is using data to drive actions and decisions, data becomes as valuable as the business outcomes it is used to drive. Data should be viewed through the same lens as any other asset within the organization: with an eye towards leverage. It’s a good idea to place a valuation on your data, if only to underscore to everyone in your organization the importance of bringing that vast quantity of data to full effect.

·    Data should be democratized. To get the most out of your data assets – having the most possible people use the data (data use being the only point at which data has value) – the organization should strive to get each individual user the most actionable information to help her make better decisions and perform her job most effectively. Therefore, people need access to and the ability to use the data. As such, the organization must cover the costs of ensuring that data isn’t misused, including data management practices, training and education, and appropriate tools across the data chain. This investment in democratization reduces the business costs of poor and untimely decisions and actions by removing bottlenecks in the flow from data to action.

·    Self-service is a goal because end-users rule. The democratic use of data only outweighs the cost of managing the distribution of data assets if the data is actually used. To be actually used, an end-user needs to feel confident and comfortable in his ability to get information that is interpreted, visual and easy to understand, and as on-demand as is practical.

·    Automation is foundational, but isn’t the end game. To get enterprise data to your teams in effective ways, a lot of data has to move and be interpreted. That kind of scale requires humans be removed from the data flow in places where automation and scalable machines and software can make a difference. Infrastructure management, data management, and even analytic methodologies themselves are all candidates for automation. This leaves “less-scalable” human beings to do the things people are still better at doing than machines: asking good questions; conversing with and convincing one-another; weighing tradeoffs; making decisions in places where data, strategy, and values all meet; and executing on complex business initiatives.

·    Technology should be simple and easy to maintain. While end-users drive data provisioning and technology demands, they cannot be the only consideration if scalable automation is to be achieved. End-users will also have new and changing demands, and the business cannot wait for custom solutions every time a new need arises. The organization should endeavor to keep technology stacks a) interoperable through stack loyalty (as long as it makes sense to do so), b) limited (not buying every new shiny object), and c) updated (to avoid interoperability collapses and downtime). As your business becomes more and more successful, your analytic questions get more complex. More complex questions usually demand connecting data across data silos. Mitigating the risk of not being able to connect data because of downtime and/or interoperability constraints becomes more and more fundamental to your business’ ultimate success.

·    Curiosity is the cultural currency of analytics. Analytics doesn’t finish answering questions; it brings you to a better set of next questions. Data is the underpinning of a “Learning Culture,” and a Learning Culture is driven by curiosity. When people ask questions about their business, they are open-minded enough to improve their business. Analytics helps frame an evolving set of questions that endlessly seeks to make connections across your market, your customers, your business operations, and your desired outcomes.

·    Reporting is not analytics. If someone is curious, that someone needs time, tools, and techniques, to answer her questions. You cannot spend time asking the next curious question if you are too busy building reports. Once you answer a question, you may seek a standard report to check in on that question again and again. That becomes a job that can be automated. Automation of that job drains the reporting work from the analytic team so that they are free to move on to the next curious question. If you don’t have a way to drain reporting from analytic tasks, you will eventually fill up the team with reporting tasks and stop doing analytics; stopping analytics stops the “next set of better questions,” and you’ve now hardwired stagnancy into your organization.