Dig deeper than the initial store's average, which is good. In this video, learn that the individual branches have high volatility—some are good, some are not efficient.
- [Narrator] Now, put yourself in Sharon's shoes. You'll find the data for this case in the exercise file "05_03". The first calculation you see is a company-wide conversion percentage for WearOne. Across all 15 locations, WearOne converts, on average, 17.6% of the customers that enter its stores into purchasing customers. This is slightly higher than the industry benchmark of 17%. Now if you stop the analysis here, you might think WearOne had no problems with its conversion percentage metrics, because overall it is better than the industry benchmark.
However, recall that an average is only one metric of the central tendency of the data. It does not represent the actual result for any individual store, and in this case, It's not telling the whole story. Take a look at this graphical representation of the WearOne conversion percentages for all 15 stores relative to the average conversion percentage. Visualizing statistics is an important way to understand and process information. This graph rank orders the stores' customer conversion percentages from lowest to highest.
The chart shows the individual store conversion percentages are generally far from the mean. In fact, only a single store, Pittsburgh, Pennsylvania, is near the average. while more than half of the stores are below the benchmark value, only a third of the stores are well above the benchmark conversion percentage. The median store has a conversion percentage of 14.1%, meaning at least half of WearOne stores are in the problem zone, below the industry benchmark, and the data exhibits quite a range of values.
The minimum and maximum conversion percentages for any store are 10% and 35.9%, demonstrating a range of over 25%. So, it's obvious that some stores are doing better than others. These stores could help the underperforming stores by spreading knowledge of their best practices. And generally speaking, decisions in the ability to implement changes occur at the store level. Therefore, the analysis of conversion percentages company-wide were unhelpful, and not actionable.
Instead, it generated a false result of overall better than average performance. Another issue emerges when you ask yourself which stores are at the bottom of the conversion percentage list. Miami, Denver, and Los Angeles combined are some of the most visited stores. This means that your particular problem areas are in stores that could really influence overall company performance, on a going-forward basis. Also note that the top three performing stores are in relatively smaller locations: Albany, Hartford, and Portland.
Make sure to take into account how that would impact the conversion percentage before moving forward.
- Qualitative vs. quantitative data
- Data analytics success stories
- Making predictions
- Asking the right questions
- Collecting data
- Understanding averages
- Sampling: pros and cons
- Cause and effect