VP Information, Data, And Insight. A key question of analytics raised during a recent presentation. What is the distinction between evidence, knowledge, and insight? This seems to be an odd topic for people who work in the analytics field since we are so fluent in these terms. The notion that other people don’t get it’ sounds peculiar to all of us. But the fact is that the general lexicon often uses these phrases interchangeably. These questions are also incredibly true and necessary to answer.
Data are the raw numbers we get under some criteria. It is very important to have standardized guidelines. As data can be incredibly difficult to record according to various standards. The old question, for example, is how long is a string? The approach depends on which unit of measurement you use. We will find such numbers if we use the Metric system. This can vary depending on whether we use meters.
Even so, the concept of criteria we implement is one of the most critical steps of any research effort. One of our first activities when doing analytical projects is to review the existing data structure of the customer and standardize this data. We ensure that all the like is evaluated in the same manner, in other words.
Information is a set of data points that we can use to explain the calculated item. Going back to our example of this string, let’s say, uh, 100 pieces of string generated by our company. We decided to quantify it in centimeters and log each of these string pieces in weight. A data point is considered for each calculation. But, taken uploaded information presented with a range of really positive information. For instance, if all the string pieces are 10 cm long, however many are not, we know that something is not exactly right in our process and we probably have to act.
In order to comprehend what happens in this scenario or pattern, we assess insights by the study of facts and evidence. The experience will also be used to render strategic choices easier. Moving back to our string parts, we need to analyze the data to decide what behavior our string parts are not what we expect. So if we see that for example, our medium (i.e. average) string length is 9.5 cm, we know now that we prefer to have strings shorter than we like. We need to know how consistently our method generates these string parts to make the right business decisions. In order to figure out what difference there is in the results, and how far our outputs are upper and lower. One approach is to look at the standard deviation, which shows the relative variance between different variables and the average.
Now we can look at our top and bottom limits and see our longest and shortest series. Let’s assume that our longest piece of string is only 9.75 cm and our shortest only 9.35 cm, which means that we are very likely to cut the pieces too short in our production. We must monitor our computer and adjust where it is and where it is cut. However, since not all string sections are the same size, we know that there are other problems which can push away that we don’t want to do. To counter these, we have to obtain additional data after we update it, compile it in new data, evaluate it to gain new knowledge, before we can make the strategic decisions necessary to drive our target results.