It’s a Data Filled World
We are surrounded by data everywhere. What is data? Data is essentially pieces of information. Just for a moment stop and think about the information that we received on our way to work every morning. We may not realize just how much information is bombarding us. Probably because we have been conditioned to the bombardment of advertisements, direction signs, random chatter, etc. We would literally filter out anything that is of no use to us.
With so much information surrounding us, we need to be able to filter what is not useful to us. This is true when it comes to collecting and analyzing data as well for businesses as well.
It’s nothing new for businesses to use data to improve their processes. In recent years, Big Data has become increasingly popular and important. As the name suggests, Big Data involves a lot of data. The more data we received, the messier it gets. To make use of the data gathered, we would need to filter and analyze the information to gain Insights into the data.
To make things more efficient it would make sense to ask relevant questions before gathering a whole load of data. So, we know what we are looking for:
- What are we looking for?
- How can we get the data?
- How can we translate the data into Actionable Information (Insights)?
What are we looking for?
Before we start anything, we first need to define the “Why” of what we are about to do.
If we do not know why we are collecting data, we would end up with information that doesn’t help.
Defining the Why would first require us to define the problem and what kind of solution we are looking for. For example, if we want to know why Store A is performing better than Store B in terms of sales, while both locations are identical.
Problem/ Issue we are looking at:
Poorer Performance of Store B
Solution we are after:
How to improve the performance of the store’s overall sales.
Method:
Compare Store B’s service level to that of Store A. What is Store A doing better?
That gives us a clear definition of the issue we are facing and the solution we are after.
Let’s take a look at something on a larger scale. What if a multinational company wanted to check the service levels in all the regions it operates in.
Problem:
Benchmarking the Service Levels of stores in different regions.
Solutions:
Coordinate a project global audit project.
Method:
- Benchmark regional performances to Global Overall Average performance.
- Compare regional performances.
- Zoom in and identify weak performing areas.
- Identify trends and week performing stores that did not comply with company service standards.
How can we get the data?
Now that we have a clear definition, we can start to look at how we can gather the data. There are many methods on how to gather the data needed. Companies can conduct internal audits, get the staff to fill in surveys, etc. Regardless of the method, all methods have pros and cons to them. For example, if a company were to conduct staff surveys, the question would be “how accurate is the data?”.
In the case of the example in “What we are Looking for?” where we would like to know the difference in staff performance for Store A vs Store B, we can use a Mystery Shopping Program. The data gathered would then be entirely based on shoppers/ customers point of view.
This method also allows the company to plan on how long a project will run, e.g. 1 month, as well as the sample size e.g how many customers to partake in the program. Which will allow for more variations in the data gathered. Instead of having gathered in a single day or week. This would allow for a better-aggregated picture of what is happening.
Of course, there are many other methods of gathering the data. The questions that arise are:
- Can the method gather the correct kind of data?
- How accurate is the data?
- Does the data reflect the scenario or the issue we are facing?
How can we Translate the data?
We now have the data that we need. How can we translate the data in a way that will help us?
Let’s keep with the Mystery Shopping Method example. The shoppers that conduct mystery shops will submit surveys based on their experience. The surveys are very structured and objective based, which makes it easier to collect the data needed.
Considering that everything is structured we can draw conclusions from an analysis of the dataset. The structured and objective based nature of the surveys will allow us to easily compare sections or questions.
We can see from surveys which section a business is losing points. For example, if we were evaluating a retail visit the sections involved might be Welcoming the Customer, Finding Out Customer’s Need, Product Recommendation, Objective Handling, and Closing. This is just a basic structure of how it would look like.
Another good point is that each question and section is scored. In an instant, we can see which sections performed better than others. So, if we see both Finding Out Customer’s Needs and Product Recommendation scored low marks, we can conclude that the staff needs more training in asking the customers questions to find out what they are looking for. This will, in turn, allow the staff to give better recommendations.
We can dive into the problem even more deeply by asking more questions. Are the staff not being passive? if so why? Do the staff know what to ask the customer? are the staff not confident in their product knowledge? etc.
However, for the most part, the data gathered using Mystery Shopping programs would be more quantitative. Which is very useful for answering many questions. But, what if we needed qualitative data?
Because of this most Mystery Shopping surveys include open comments that allow shoppers/customers to write their thoughts and opinions on their experience. This too has a slight downside. If the sample size is 10 – 100 it wouldn’t be too difficult to read through the comments. however, if the sample size is 1000 and above the task would be too tedious and labor intensive. It may not even be very fruitful to go through all the comments, which makes it less efficient and a waste of resources.
Insights Analyst would pick interesting comments to share with the client. This is done when the Quality Check team is going through the surveys. Analyst would also filter the sample size to those that performed well and those that performed poorest, collecting qualitative data from those surveys. As this would show the shoppers/ customers opinions on what happened during the visits.
After gathering and analyzing all the data. It would then need to be presented in a way that makes sense to a non-data-center reader. This is where Data Visualisation comes in. Visually presenting the data in graphs and charts, with notes on analysis. The full analysis would be compiled in a Management Report and presented to the clients.
It is then up to the client to use the data that is gathered, analyzed and translated into putting into action processes for improvement e.g. training on specific areas of customer service.
