Most organizations tend to "push" information from process step to process step, whether the subsequent step is ready to do anything with it or not. People within the process are pleased just to "get it off their desk," as the work becomes "somebody else's problem."

"Push" environments tend to give rise to a high level of expediting, and priorities are made based on "who yells the loudest" or "the squeaky wheel." Further, it creates functional "islands" where work is "thrown over the cubicle wall."

Such practices result in outcomes counter to the objectives of the organization, and the people who work in them. Rather than getting more work out the end in shorter time frames, the opposite holds true. Bottlenecks arise, lead times increase, throughput decreases, and quality tends to decline. It is analogous to pushing on the end of a rope or chain -- the rope or chain gets bundled up, gnarled and ugly. This is just what happens in any business process. So what is an organization to do? What is the alternative? Lean Thinking tells us that we should "pull" the work through the process rather than "push."

Pull is defined as a method of controlling the flow of resources based on actual demand or consumption. In its most basic sense it is a decision tool. In manufacturing, it helps organizations decide what to make and when to make it, and what to buy and when to buy it. In offices and services the resources that need to be controlled are information and people. The same pull concepts used in manufacturing can be applied to office and service organizations. They can help office and service workers decide what to work on and when to work on it in order to maintain customer service, while preventing overproduction.

In general, it is a Sequential Pull System that finds the most frequent application in an office or service environment. Here a limit is placed on the queue that when reached triggers a decision. In the case of a maximum limit, it could mean that the supplying process (i.e. the "supplier" of the information) decides to stop processing a particular type of information and begins processing a different type of information. There is no need to continue to process more information beyond the desired maximum (i.e. overproduce) as the downstream or customer process cannot process it in a timely manner. The "supplier" might as well work on something else. 

In the case of a minimum limit, this would signal the "supplier" to return to processing a particular type of information because the downstream process will be ready for it. One can easily see that the flow of information as well as the "supplier" can be effectively controlled by the establishment of such basic decision rules. Further, the "supplier" can self manage -- requiring little or no direction from a supervisor or manager.

At a design-to-order company where every order requires some level of design effort and other activities to be performed prior to production (e.g. estimating), the information was a hardcopy "project" placed in a folder. A series of baskets hung in visible locations (a central aisle) in the office provided queue visibility. The baskets were timed in such a way that anyone would be able to identify folders that were not being processed in the timeframe established. This was an indication that one segment of the order process was having difficulty keeping up with demand at that particular time. Over time, associates in the office learned to respond to the visual signals provided by the folders in baskets with little or no direction from the office manager. In other words, the pull system became "worker managed."

In a Sales example, a company previously had what could be described as a "sales funnel" where pre-qualified sales opportunities were placed in queue to be followed up upon at a later time. These opportunities were identified by geographical region -- in reality there were multiple "sales funnels." As part of a lean office effort, the organization made these queues visible. They also established goals with regard to the lead time in which they wanted to follow-up on these sales opportunities. Once this visibility was provided it became apparent that particular regions were not able to get to their opportunities in a timely manner. Sales management would monitor the regional queues and re-direct particular opportunities that were approaching the established lead time goal to sales associates who had available capacity. In other words, sales resources were "pulled" from other regions to meet demand in another region. This was a major contributor to the 12% one-year increase in revenue.

View article on one page