To help visual the idea, imagine you have just arrived at Boston 'Logan International' Airport. You have been traveling all day to get to the city, you claim your luggage (if you are lucky) and set off to find a taxi to take you to your final destination. You walk outside in the below-freezing temperatures to be elated to see a line of available taxis just waiting to load happy passengers. Then you see the queue of unhappy passengers waiting to get into taxis. Two airport staff are coordinating the flow of passengers (presumably to prevent anyone walking in front of a moving vehicle), and the flow of taxis (checking licenses and handling special requests for minivans, etc). Still, the people and the taxis are both stood in a queue for 20 minutes waiting to be united. Something is wrong.
In this example, the staff coordinating the flow of people and cars are like the supervisors dishing out work in an office and managing the workers tasks; the taxis are the workers; the passengers are the work cases to be handled; the taxi-rank and the road out of the airport is the workflow. Only one of these is the issue - the taxi-rank / workflow. How is this so?
In our example, the passengers are not at fault for wanting taxis, or occasionally minivans, as much as a recruiter is at fault for submitting candidates to an HR department's recruitment 'workflow', even if sometimes a person has perfect skills but needs a special license to be employed. The work exists and needs to be handled, even though at times there may be a valuable, albeit complex exception case.
Again in the taxi-rank example the taxis are not at fault for waiting in an orderly line to enter the taxi rank. They are doing what is expected of them, for the hope of suitable compensation. They are available for work. The same thing could be said of a team of temporary workers in an office tasked with handling the flurry of new job applicants when a company is hiring for multiple new positions. They are available for work, and will do it if it is available. If not, the coffee machine probably needs refilling regularly until the work is handed to them.
Finally, within the constraints of the current taxi-rank workflow, the staff coordinating passengers and taxis are also doing nothing wrong. Nobody is getting run over, taxis are correctly licensed and drivers are courteous to shivering passengers. This is the same as the supervisor taking some new resumes and handing them out to the workers as soon as the supervisor sees they have finished the last batch, then collecting and checking the results. Work and passengers seem to get matched with workers and taxis just like clockwork.
The problem is there is still a 100 person long line of passengers and 500 resumes to be reviewed, despite the apparent efficiency of the clockwork production line workflow. The taxi-rank / workflow itself is at fault.
At the airport, the taxi-rank forces all available taxis down a single orderly, safe, easy to manage lane. This results in holding up 40 idling taxis while 7 are loaded with passengers, one passenger finds that a pair of skis are too large to fit in a taxi and needs to request a minivan, and 90% of the passengers wait in the bitter Boston winter weather. The recruitment workflow similarly provides the appearance of orderliness and efficiency, but is really a choke on work getting assigned or handled as the supervisor hands out new resumes and collects results of completed work. There is always time for another coffee between batches of resumes landing on a worker's desk, especially as little brain-power is requested to do the repetitive work. The workflow is leading to inefficiency and a long wait for work to be completed.
How well does the taxi-rank or production line workflow respond to a change in demand or volume of work? Imagine the result if a new plane arrives at the airport, or a new job is posted for immediate placement; there is no way to make the workflow go faster. It still ticks along at its own pace with a long queue of work to be done, as taxis shuttle in and out, and resumes get delivered to workers' desks. No amount of supervision can make a well structured production line go faster.
Surprisingly, a reduction in demand with a proportional reduction in workforce is even worse for the production line workflow. When work enters a production line sporadically or at a rate lower than it was designed to handle, the work actually trickles through the end to end process slower. Proportionally fewer resources handling work leads to proportionally longer lags between sequential tasks being completed and the work stutters along rather than flows. The recent economic crisis and its impact on car manufacture has sadly demonstrated this: it was better (for companies) to shut factories for weeks on end, rather than trying to run them at half capacity. The efficiency experts at these companies probably recognized that it would cost way more than half to achieve half the output of vehicles.
Unfortunately many organizations have locked themselves into a restrictive production line way of working in the office, when truly there should be far more flexibility. From the inside, to the supervisor it may seem like there is no way out. Often the only answer to remove the log-jam is a change of tool such as a new workflow system, or an outside view to design a new workflow on top of the current tool. New software or a fresh perspective maybe all that is required to make things work better.
With a new year hear, what better time to break the old, inefficient habits of the past?