Guest post by Kurt Peckman, Program Director, IBM Business Analytics
Follow Kurt on Twitter: @KurtPeckman
I mentioned in a previous blog that I commute on the train to work.
A week ago Monday I found myself in the middle of dispute between the conductor and an irate female rider who had no monthly ticket and no money to buy a one-way ticket from the conductor.
The first day of each month often generates unnecessary stress on the train for monthly ticket holders. That’s because the chances (e.g., the probability) of having and presenting a valid monthly ticket on the train on the first day of the month has to be lower than at any other time of the month.
Until recently, Metra (my mass transit provider in Chicago), granted passengers amnesty through noon on the first day of a new month. However, in an effort to close a budgetary gap and to tie strategy to execution, Metra implemented a simple and effective, zero-tolerance business rule that provided instant ROI: all riders must now present a valid ticket when riding regardless of the day of the month. By implementing better governance of operations through its conductors Metra was able to easily increase revenues.
This brings me to the start of last week – Monday, Oct. 1 – when analytics and decision management (and general bedlam) exploded on the train.
The rider, clearly embarrassed and annoyed, was pleading her case to the conductor about not having her monthly ticket and being broke to start of the week. I could empathize because I’ve been in that position many times myself.
The conductor, implementing the new zero-tolerance policy, has a limited number of actions:
– Grant situational amnesty and run the risk that this woman is lying and may do it again, thereby setting a dangerous precedent with eye witness passengers that the policy is really not zero tolerance;
– Upon arriving at the station, accompany the rider to the ATM then to the ticket window, watch her buy a ticket, then punch her ticket with a stern glare; or,
– Throw her off the train at the next stop.
**Note: While not very customer-centric, options #2 or #3 would definitely cause a spike in compliance.
My point, however, is that Metra is still leaving money on the table.
Having conductors dedicating precious time in transit to riders with no ticket and no money is not optimizing revenue generation. The resource constraints facing every conductor – that of processing an entire train while in transit (where “processing” means confirming and selling the occasional one-way ticket) is tough on the first day of any month and next to impossible when the first day of the month falls on a Monday.
An IBM Analytical Decision Management approach to this business challenge might look like this:
· Dedicate a single sales agent to selling one-way tickets on the train during the first day of the month – especially when that first day falls on a Monday. This agent would only be responsible for ticket sales. Notice the optimized crew size.
· Give this dedicated sales agent a tablet that is hooked up to various Metra databases like the monthly “ticket by mail” or “regular customer” database. Doing so would enable an easy way to confirm the semi-legitimate excuse that “my monthly ticket is at home” or “I forgot today is October 1.” Note the critical role of business intelligence data.
· Upon confirmation that the rider is indeed a regular, fair-paying customer (a “score” that is based on business rules, predictive segmentation, or both), update his/her information in the database to trigger a form letter or reminder next month that explains how ticket prices can be kept down only through high levels of compliance.
· Personalize the reminder message for each individual rider using predictive analytics to improve the probability of the message’s impact (e.g., “If you forget your ticket again I’m going to tell your wife/husband/mom/BFF/parole officer how irresponsible you really are.”)
· Ask the riders that are not in any database and who have no cash to pay for their one-way ticket electronically (via the tablet). The cost for this ticket would naturally include an additional fee to help pay for the new “Ticket Agent with a Tablet” or “TAT” program.
In this example, business rules, predictive analytics, optimization, and scoring (all the elements of a Decision Management solution) have come together to improve the customer experience and increase ticket revenues. Optimized decisions have been automated through a mobile device to enable a single ticket agent to link strategy to execution rider by rider – making sure that every single rider has actually paid something to be on the train.
Metra is but a microcosm for many, many businesses where making the right decisions and implementing optimized day-to-day operational actions on a case-by-case basis are critical to organizational success.
On your way into the office tomorrow – even if that trip is from the kitchen to the study – consider all the potential organizational success and improved customer experience you would experience if more of your “providers” leveraged a quality Decision Management system. For example, imagine if your network connection speed was based on your job title (rule), forecasted bandwidth need that day (predictive), and network load (optimization) or if the discount offered to you at the coffee shop was based on your loyalty program segment (rule), probable lifetime value (predictive), and supply of coffee beans in stock (optimization).
As for the standoff on the train, things ended well when I offered to pay for the rider. Remembering me and understanding the gesture, the conductor backed down and asked the lady to purchase and present a one-way ticket on the ride in tomorrow. The rider dropped her attitude and agreed to purchase and present the requested ticket and Metra maximized revenues with this rider.
More importantly, Metra prevented the need to throw (someone’s) momma from the train.
For more information:
· Watch the video and learn how to maximize results by automating and optimizing decisions