Last year I wrote about some of the lessons that can be learned from a music festival called Floydfest in the Appalachian Mountains near Floyd, Virginia. This was illustrative of the more general point that we should always be learning and learning should happen everywhere. The festival was a role model for many business concepts (http://jeffglass-engineeringmanagementblog.com/2012/08/). This year, not so much. Of course the situation did not fundamentally change from last year to this year but some external events and internal decisions conspired to make the festival a real challenge this year. But the point is still the same as my blog post last year – you can learn from every experience! So what happened?
If you live on the East Coast, you know that the weather this past spring and early summer was not typical. The amount of rain we had was much, much greater than usual. As such, an outdoor music festival probably needs to consider this as a potential issue and have contingency plans for handling the crowds during a rainstorm. And to be fair, I am sure that Floydfest leadership did consider this as it has rained many times during the festival in the past. Thus, achieving safe electrical power in a rainstorm, stages that can protect people and equipment, and temporarily halting of the festival all seemed to be considered and utilized. But something that might be loosely considered a local “Black Swan” event occurred at this year’s Floydfest and it was not in the contingency plans. The concept of a Black Swan event involves an ultra-low probability event; an outlier. It is an event “that is unpredictable yet has wide-spread ramifications. Not only are Black Swan events difficult to predict, but [Nassim Nicholas] Taleb also argues that we human beings have certain psychological limitations and biases that prevent us from foreseeing these events, while also thinking that the events were perfectly predictable after they occur” (http://www.black-swans-explained.com/). There are many articles about Black Swan events so I will not try to cover them here (as recently as the terrible floods in Colorado last week and as general as “rogue waves”; phenomena have been described by the Black Swan concept). If you don’t know about Black Swan events, start reading up on them because every manager should be aware of this concept (rogue waves are pretty interesting too so you might want to check them out). My interpretation of this in an engineering management context is that we need to be careful about relying on our analytical tools when predicting the probability of an event. This is not related to the tool itself but, as always, it is the assumptions we use in the tool. For example, most of us have experienced a very surprising coincidence in our lives. Perhaps you meet someone from your home town on the other side of the world in an airport. The probability of such an event may seem vanishingly small, entirely unpredictable. However, if we consider the total number of possible encounters across the entire population and the amount of travel the entire population participates in, it is clear that someone at some point in time is going to run into someone from their home town somewhere on the other side of the world. It is just a matter of time. So when planning for the future as an engineering manager, one must consider not how to predict what ultra-low probability events will happen but rather are your processes robust and flexible enough to overcome a Black Swan event? Since the event is not predictable it is not useful to conduct scenario planning and try to have the right contingencies for the event. So how is all this related to Floydfest?
In the middle of the festival, on top of a ground that was already saturated and a water table that was likely higher than it had been in decades, a monumental rainstorm struck the area overnight during the festival. It rained for hours and turned fields being used for parking into literal mud-pits http://www.roanoke.com/living/music/2107669-12/rain-wreaks-muddy-havoc-at-floydfest.html). Even walking at the festival became impossible with mud up to a foot deep on the main walking path. In between almost falling multiple times and losing a shoe to the mud, I took the opportunity to do a little research during all this – i.e., I went people watching! But the main problem was the parking lots. Cars were stuck everywhere and numerous accidents littered the parking lots on festival grounds. Near misses of cars sliding into pedestrians made the situation undeniably dangerous. So leadership of the festival did the prudent thing; they shut down all parking near the site (including handicapped parking and prepaid expensive VIP parking). They began busing people to and from very remote (but relatively safe) parking lots up to an hour away, not including the wait times. The situation was so dire that volunteer workers were taking the buses into town to go to Laundromats to clean and dry their sleeping bags because their tents had filled up with mud! I spoke to some people who wasted more than 3 hours driving to the remote lots and waiting for buses to bring them to the festival. VIP customers that paid hundreds of dollars for special parking and closer access to the performers were being turned away from their usual parking areas and being told to drive up to 60 minutes away and then take a (yellow school) bus back to the festival were confused and frustrated (OK, they were angry!). What went wrong? Was it just bad luck or could it have been handled better?
If the rain and resulting “mudfest” was not really predictable, if it was a local, let’s say “mini-Black Swan,” event, what could the organizers have really done differently? It is all about process. That is, if the actual event is unpredictable then your processes must be set up to react quickly, flexibly and effectively in the event of a crisis. I think this also implies that you need to have a way to move from a decentralized empowered management style to a centralized, command and control style in short order. Exactly how centralized you need to be depends on the scale of the Black Swam event. If the entire festival is experiencing the crisis then control needs to go to the top; to those with the visibility to see the entire picture, from the venue to the artists to the vendors to the attendees. This change requires GREAT communication and clarity about how decisions will be made in a crisis. Your front-line staff that have the most interaction with the customer need to know what is going on, how their jobs have changed and how to respond to questions. I noticed that when staff were clear and empathetic, the customers were reasonably satisfied, even when being told very bad news. When the employees had an attitude of: “good luck, we don’t have any idea what is going on either” (yes, that is a quote) then the customers became rather belligerent. Staff had no directions to the new locations for parking, no idea which buses to take to get to the right parking area and in some cases, no idea that different parking solutions even existed! So it seems that the leadership did not disseminate the decisions effectively and did not train the staff how to manage customers in the face of bad news. In fact, why didn’t the leadership also communicate directly with the customers when they determined that the situation called for drastic measures? They had emails for all of the attendees and many of us were finding ways to check email during the concert (including a VIP booth just for that purpose and a special internet company tent that gave free Wi-Fi access that was full of attendees throughout the festival!).
In summary, analytics are great but they cannot predict everything. Understand that extremely low probability events can and do occur. Especially in the world of technology and engineering we should not fool ourselves that probabilities tell the whole story. Black swan events do happen. They are not predictable but you can still prepare for them by being flexible, responding quickly to your environment and developing a process for how to make and implement decisions in a crisis. Great communication will also be critical.
