Analytics in daily life

Analytics is not a tool or a technology; rather it is a way of thinking and acting. Most people use analytics to resolve problems faced in their day-to-day life, mostly without even realizing it.

Imagine a simple situation – you need to buy a new TV.

Regardless of your budget, chances are you will find more than one option available in the market and will need to select the one that best fits your need. Aside from the various manufacturers, you will need to choose between a host of features.

Flat screen vs. Conventional?

Plasma vs. LCD vs. LED?

Screen size, audio quality HD readiness and so on.

, Analytics in daily life

The first thing is to simplify this clutter of information. Out of all the features being hyped by various brands, which ones are really important to you? Is HD readiness important in your TV? Can you really make out any difference between a refresh rate of 60Hz and 120 Hz?

Comparison of attributes

Let us say you narrow down your list to 4 features – screen type, screen size, audio quality and power consumption. You will now rank these attributes in terms of their relative importance to you.

Maybe screen type and size are very important to you; audio is of medium importance whereas power consumption has a low significance barely making it to the short list.

You will then evaluate the set of TVs that fit your budget on these attributes and score them based on the importance of the attributes. For example, a TV with a large screen is more attractive than a TV with low power consumption.

In a sense, you are creating a table

Implicit or explicit?

Most likely, you would never actually explicitly create such a table. Rather this process of short listing the attributes and grading the models would work sub-consciously in your mind. For example, the sales person’s comment about the light weight of a particular model may largely be ignored by you whereas the piece about 6 integrated speakers may make a significant impact.

Let us examine how the problem of buying a TV was broken into simpler parts which could be answered using information backed by data.

  1. Short-list the attributes important for you in a TV
  2. Rank order the importance of attributes
  3. Select models that fit your budget constraint
  4. Score the selected models on the important attributes
  5. Choose the model that best meets your requirements

Simple problem

This is an example of using the approach of analytics to solve a real-life problem. A complex and abstract problem is broken down into a series of smaller problems that can be solved using the available data. This is a relatively simple problem where the process of analytics is happening implicitly in our minds. There are certain aspects which make this problem relatively easy to solve. The number of models being compared is 3. The number of important attributes is just 4. The process of ranking the models on each of the attributes is relatively straight forward as the required information is easily available. The limited number of models and attributes as well as easy access to fact based information makes this a mental problem where sophisticated statistical techniques are not required.

Increasing the complexity

Now imagine, you have a million different models to choose from. You have to select not one but 10000 models. There are hundreds of attributes and you are not really sure which ones are important. Now the problem suddenly becomes a lot more complex. This situation is akin to the questions businesses face every day in order to run and be successful.

Real life business problems

Say you work for a credit card company. Your company has tied up with a leading airline to offer upgrades to customers who use the credit card to purchase their tickets. You plan to send an impressive looking brochure with the offer details to some of your customers. You can’t send it to all your customers because there is a cost associated with printing and delivering the brochure to each customer. However, there is promise of a higher payout from those who do respond. You plan to send these brochures to 10000 of your customers and you would like to identify those who are most likely to respond so you can maximize the return on this investment.

How do you decide which 10000 to choose? Now the problem has become similar to the situation mentioned earlier where you had to select 10000 models from a million.

Interested in learning about other Analytics and Big Data tools and techniques? Click on our course links and explore more.
Jigsaw’s Data Science with SAS Course – click here.
Jigsaw’s Data Science with R Course – click here.
Jigsaw’s Big Data Course – click here.

Related Articles

} }
Request Callback