Big Data. It’s the new thing. Alright, it’s not so new. But it’s the thing. Even “small data” would seem to be useful. So why are most data not used, even if they are collected?

Have you ever heard about how important it is to put measurement systems in place, and then seen how much time and money is spent implementing data analytics systems, only to find later that people are just not using them? The most common reasons include:

• Too much raw data – people are overwhelmed by the sheer volume of data;
• Lots of useless data, mixed in with some potentially useful data;
• Lack of clarity on what the data mean, and what to do about it.

Here are some basic definitions that can really help in making data and analytics useful:

1. Data (which is the plural of datum) – just raw numbers;
2. Information – data that has been sorted and organized;
3. Analytics – trends and comparisons that can be drawn from the information;
4. Insights – actionable conclusions based on an understanding of what the analytics tell us.

There are several reasons we want to use data, information, and analytics. First, we need to know:

• what happened? Then, we need to understand
• why did it happen? Then we need to draw conclusions, such as
• was this good or bad? Finally, we need to know
• what can and should we do about it?

Because, in this digital age, we can get overwhelmed by data overload so easily, it is better to start small by asking the above questions in very simple terms. Begin by picking a few key variables, then put the following plan in place:

1. Instrument your environment (put the tools in place to measure the key variables that you want to track);
2. Collect the data for a period of time (whatever period gives you enough data to make a reasonable series of analyses and decisions);
3. Format the data into information that normal people can read (you don’t want to require a Ph.D. in statistics in order to make sense of the information);
4. Analyze the information for that period (create some graphs, look at the trends, see what the basic information yields);
5. Develop conclusions (try to understand why the trends look that way, or how come the number of users on mobile devices is lower than you expected, etc. – these are the insights that are the ultimate purpose of having the data in the first place);
6. Develop a plan of action (create a series of steps that you can take to modify what you are doing so that you can achieve the results you want);
7. Implement these changes, then go back to step 2 and run the experiment again.

Once we have used this process to create a continuous improvement cycle, we can further develop our analytical skills to predict performance in the future. But that is a topic for another day…..