Kill the noise with data - A BigData Use Case

Maximum noise is produced by the exceptions. Managers beware.

The Pareto (80/20) principle holds good at all times. Period. I heard this incident (now an anecdote) from an ex-CXO at an erstwhile taxi aggregator startup and found it so interesting that this story ought to be told. So often than not, as management we are swayed away by the noises created around us. It’s quite ironic that these noises that compel us to take life altering decisions are more often than not produced out of exceptions.

Just the other day I requested my new intern to call up some of our existing customers to introduce a new product and get their feedback. On inquiring, she reported that one of our customers seemed extremely unhappy with her call and backfired her with complaints like when the supply of our current products is insufficient to them, how can we think of introducing a new product and she went on offloading her distress upon me. 

Once she was finished I quietly inquired how many other customers had she called. The answer was 9 others. “And what was their reaction”, I asked. “Yeah they were quite happy with the new introduction”, she replied. Well, there you go – a live example of how the exceptions cause the most ruckus.

Thus it becomes necessary for managers to carefully identify and ignore the exceptions to be able to see the big picture.

This other incident narrated to be by my friend will help me rest my case. A big problem arose before the management team of this erstwhile taxi aggregator startup based out of Bangalore when their sales and marketing team started showing dissatisfaction towards the one-fare-for-the-entire-city policy irrespective of a particular region’s proximity to the airport. The insistence was that the fares should be reduced for the northern region as it is closer to the airport in comparison to the rest of the areas and this will help them satisfy the customers in the north. The management was under pressure to give in to the demand and introduce differential fares. However, they chose to look into the data before taking any decision. 

After comparing bookings from all four regions (north, south, east and west), it was revealed that the north region (closest to the airport) had close to 0% contribution to the total bookings while the south region (farthest from the airport) had the maximum i.e. close to 80% contribution to total bookings followed by the east and west respectively. So, the sales and marketing insisted that they want the reduced fares for the north in order to increase the booking contribution from the region. However, the management stumped them by presenting an analysis of the demographics of the north region (showcasing the income group residing in the area) substantiated by booking patterns from competition (all of which had minimum bookings in north) thereby establishing the fact that due to its low income residents and proximity to airport, the north region can never be a heavy contributor to business hence it is best to concentrate on rest of the regions and keep the fares unaltered.

Thus we see how analytics has become a way of life (business) and data is important to kill the noise.