The Model T of Analytics
This is Part 2 in the Driving into the Unknown of Analytics Series. Part 1 covered the analytics community and how it started as artisans. This part will cover how businesses started to make an assembly line of insights.
The World on 4 Wheels
The Model T is described as the car that put the world in motion, and it is easy to see why, millions were made throughout its history. It was a reliable car, very capable, and sold at a price that was affordable to the middle class. Because of this, the Model T was the first and only car for most families across the United States in the early 1900’s. It single-handedly put the world on 4 wheels and transformed the landscape.
From Artisans to Factory Workers
The success of the Model T cannot be attributed to only one factor, but it cannot be denied a primary driver was changing how the car was made. Up until Henry Ford, cars were all made the same way, the cars stayed put and were assembled on the spot. This led to part uniformity to be low and worker skills to be high as parts needed to be modified to fit together. Ford changed this through the introduction of more standardization and the moving assembly line, where every worker would be assigned a small task on the vehicle as it moved on a conveyor belt. This innovation drastically decreased not only the time to create the vehicle, but also the skill needed for each worker.
In data, a similar revolution has already occurred. The artisan skills of a data scientist are not needed for small reporting, but instead, analysts can perform this work, creating the Model T of analytics: Descriptive Statistics.
Describing the Past
This Descriptive Statistics is the analysis of past performance to drive business decisions today. Its main use is reporting where the current status of the business is ascertained. For example, by looking at a ratio of passengers per flight for every route, an airline can make a data informed decision on whether or not to add another daily flight from Tokyo to Seattle. This can be accomplished with the Model T of the analytics world, Spreadsheet Software.
Spreadsheet Software has been around nearly as long as the modern computer has been, but they became popularized in the 1980’s with the personal computer revolution. With PC’s, the use of spreadsheets defused throughout businesses with everything from inventory management to scheduling. But the usage of spreadsheets didn’t stay exclusive to businesses long. With so many people using it at work, it was only a matter of time before they took their skill and begun to apply to their personal life. This resulted in a potential market of millions (or more) of users.
Perhaps the most ubiquitous spreadsheet software, and by extension analytics software, is Microsoft Excel™. Excel™ drives the business work and you would be hard pressed to find a business that doesn’t use it in some fashion. Because of this, the ability to use Excel™ is now requirement at almost every office and professional career, from the intern to the C-Suite. Its power comes from its ease of use and flexibility. In just a couple of minutes, one can copy data into it, modify the data, and create graphs describing the data.
The Limitations of Spreadsheets
While spreadsheets have a lot of uses and are very good at seeing the direct effects between data points, if airlines only looked at one factor when making a multi-million dollar decision, they would fly as well as a Dodo (or Hooters Air).
For example, just adding one more flight on an existing route, an airline will need to know how that will affect every other route in their network. Every one of their 100’s of flights per day. Let’s say that an airline adds another daily flight from Tokyo to Seattle, decreasing the load factor on the other direct flights between these two cities. This effect would allow for more bookings between these two cities and is easy to calculate in a spreadsheet software.
Unfortunately, this small change could have a knock-on effect of decreasing the in-flight load factor between Hawaii and Seattle, so much that they will be flying that route at a loss now. Why? Because when they expanded the Tokyo and Seattle flight, they decreased the load on the Tokyo to Hawaii flights, and ultimately the Hawaii to Seattle flights. One small change in flights caused an unforeseen impact to other flights, propagating adverse reactions throughout the network.
Analyzing this propagation is nearly impossible to see within a spreadsheet software. Excel™, Lotus 1-2-3, and other spreadsheets work best like a library. They allow of the user to organize data and see how data is affected by other factors, one at a time. Given this limitation, certain members of our community have taken notice of the desire to perform better analysis. This created the field of business intelligence and the field exploded with companies like Tableau, Domo, and Power BI being used is droves a crossed every business sector.
Last week, we saw the creation of our data community; where Data Scientists are the artisans of the data world, creating custom models by hand. But with the limited number of artisans, businesses have created a business analytics assembly line through the use of Descriptive Statistics; where low-level analytics are performed with spreadsheet software. Join me next time where we cross beyond the capability of spreadsheet software and enter the world of Business Intelligence.
If you'd like to learn more about how AI can help you on your data journey, download the 3 Key Features Brochure now!
LATEST POST : It's Not You, It's Us: Using Data for Employee Retention