How Analytics Could Become Data's 'Field of Dreams'

'If you build it they will come,' was a touching fantasy for the movie 'Field of Dreams' in which a farmer builds a baseball stadium in his field and a game of legendary players ensues. Unfortunately that won't work for those attempting to create insights from the mounds of data that organizations now routinely generate.

Most enterprises recognize that data itself, rendered without interpretation and analysis can be more of an impediment than a solution or success factor. But as the following article explains, even dashboards, algorithmic models, predictive analyses and all of the other 'tools' well-meaning data scientists conjure up to astonish and enlighten their internal and external clients may be disregarded if the perceived wisdom does not, on delivery, match the skills and goals of those on the receiving end.

The reality is that the same forces driving the exponential growth of technology and its statistical offshoots - convenience and inexpensive abundance - affect its useful application. The ability to turn information into knowledge is both data's biggest challenge and greatest opportunity. JL

Piyanka Jain comments in Forbes:

An actionable analysis comes from understanding the problem as well as the actions the stakeholders are ready to take. What are they really trying to accomplish? Make sure your stakeholders know how to interpret the data and then use it for analysis. If they don’t have the skills to read and analyze data, the best dashboard would go to waste.
For the past decade, the #1 problem cited by analysts and their managers has been the same: their team built the best possible model (read: analysis, dashboard, report, predictive model) but people are not using it. Even the internal clients, who asked for it, aren’t using it! In fact, this issue is so pervasive in organizations that 80 to 90% of the analysts in the audience report this problem.
Recently, one of my clients, the CIO of a $3B consumer technology company, shared with me that their company had created at least 1000 dashboards in last year or so, based on specific requests by internal stakeholders—and less than 5% of them are used. Crazy, eh? At least you know you are not alone if you have this problem. And good news is, there is a solution.
The solution lies in understanding the problem. Why do good models gather dust? Here are the top 4 reasons.

1. The most common reason is the solution is built without a true understanding of the real business question. What scenario is the stakeholder trying to solve for? Why are they requesting that dashboard, analysis or model? What are they really trying to accomplish?
2. The stakeholder doesn’t know how to act on your solution. Either your recommendation is unclear, not actionable or the stakeholder doesn’t have skills to turn your solution into insights and actions.
3. The stakeholders don’t trust your solution. Either they don’t buy into your methodology or your data—or they simply don’t trust you.
4. Finally, it could be that your solution is hard to use. Maybe your dashboard takes 5 minutes to log into, or the drag and drop functionality doesn’t really work, or it’s just plain hard to use. Or, in case of analysis, maybe you gave a horrible presentation—not talking to your stakeholder’s pain, not to the point, not sharing the “why” behind your solution.So how does your solution become one that gets used?
5. Use a consultative approach to first understand the need (the Real Business Question) and then reconfirm the need and the path to the solution using an analysis plan. Make sure you have the right stakeholders in the room. This ensures that you are building the solution to the right problem and your stakeholders will be ready to use the solution when it’s ready.
6. Next, make your solution actionable. An actionable analysis, for instance, comes from understanding the problem as well as the actions the stakeholders are ready to take. During the stage in the process where you brainstorm for hypotheses (possible causes of the problem) get clarity on actions that need to be taken if the hypotheses are proven to be true. If the project is a dashboard, make sure your stakeholders know how to interpret the data and then use it for analysis. If they don’t have the skills to read and analyze data, the best dashboard would go to waste. So don’t build it.
7. Build trust through communication and consistent delivery. Make sure your stakeholders agree to your analysis plan before you start executing. Then make sure to share early insights (or early prototypes) to get first cut feedback and adjust course, if need be. This process builds alignment and gives you the ability to influence your stakeholders into action based on the insights from the data.
8. Lastly, learn how to make good presentation customized to the audience. And, make sure your dashboard tool is easy to use. If it isn’t easy, no one will use it. If training on the tool is needed, make sure that your stakeholders get trained before you deliver your first dashboard/ report/solution. And then, don’t just send your report or analysis; walk them through it and help them get comfortable.