Why do analytics fail? Some will blame the technology and some will blame the business. Either way you think about it analytic projects usually fail because of a wide chasm that exists between the people building the analytics and the people using the analytics. There are a couple of reasons for this condition: 1. Language 2. Organizational Buy In. Without considering both of these things it doesn't matter how brilliant we are as data scientists and business people, our project will more than likely end up in some very pretty Power Point deck never to be seen again. We all work very hard to do analytic projects and they can have significant impact when executed properly. Let's take a look at a couple of things we can do to prevent advanced analytic project failure. It all has to do with the two water crafts below:
2. ORGANIZATIONAL BUY IN:
John why did you do two and then one. Ha! The goal of all analytic projects is to achieve organizational buy in. It is the reason why number 1. Language is so critical. If you can't speak to people in a language they can understand you are more than likely never getting on any boat or off the pier. Without the business people on board with you, your project may not get funded. Not every organization is data driven! (YET) If you look at the Fortune 1500 25 years ago and compare it to the FORTUNE 1500 today it is drastically different. There are many organizations that failed to recognize digitization and modernization. Data and analytics are the currency of the modern world. Not all organizations have the chops like Google, Amazon, and the like. They don't make empirical decisions, they make data driven decisions. We need to gain buy in from our business customers so we can help them do the same. For this reason all data scientists have to work on their HUMAN skills. We have to be superb story tellers and sales people. If you are great at communicating with the business you will get to ride on both water crafts above. (see picture) You will get invited to the boardroom and the yacht.
Above are two pictures of water crafts. (boat snobs I know that the top boat is a yacht)
John, what do boats have to do with analytics. Good question. Let's dig in. Both boats are fun and both boats have people on them. The people on the top most boat are business people. They take all of the business risk and have a lot of goals. Those goals are based around profit/loss, return on investment, shareholder equity, etc. They also make the decisions on funding your analytic projects or have direct line into the people who do. They also have power and influence to introduce change to an organization. These people can help you change PEOPLE and BUSINESS PROCESSES.
The people on the bottom boat are people in the analytic journey. They are great people and also having fun in their boat. They speak a different language and have different goals. Their goals are to develop predictive models, segment customers, classify things, and develop features. Success of the people in the bottom boat rely on the people in the top boat understanding what you are doing and saying. Data science has a language that is very new to the people in the top most boat. As people in the bottom boat we have to think about how we communicate with the people in the top boat. We have to use their language.
So try to put away your markov chains, centroids, and complex math equations and start to learn the language of your business partners. Craft messages they can understand. Also, figure out how to integrate the analytic answers into the business systems and processes so you can operationalize them. There will be another article on the THREE ANALYTIC MILES coming soon.
CONCLUSION: Analytics need to be easy. They need to be made for HUMANS. They require an interface that business people and humans use. For an example of these human interfaces and advanced analytics. Take a look here: Aster Analytic Solutions
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.