One of my jobs during high school was in a tire manufacturing plant in Memphis, Tennessee. On more than one occasion the hated and dreaded time and motion “guy” showed up to observe how I was doing my job. I never knew what the outcome of the observation was or whether the change to the four page process I performed to sort green tires was due to the observation. The job was never easier after the change. Reflecting on that time (and several industrial classes later) I understand that observation is an important tool for developing an understanding of how work should be done, but is not a tool to be used all the time. Using observation in the right scenarios and then taking steps to plan how you will observe is critical getting value for the effort needed.

Why observe? The simplest and clearest rational for using observation techniques is that users and stakeholders either don’t always know what they want or can’t always express their current needs and foresee their future needs. Therefore a new set of eyes will expose more and different needs. There are four typical reasons for observing should be considered as a tool gather knowledge and requirements.

  1. Physical location is a determinant. Processes and work flow is often affected by the physical location. When the physical layout of the people or machines could strongly affect the solution the team developing the initial backlog should observe the process in action to understand the nuances of flow of work.
  2. When people can’t tell you. Occasionally the process being studied will be so complex that no one is able to coherently describe how it works or how it should work. Even more occasionally asking is met by silence due to lack of trust. In both cases observation is a valid tool to develop an initial backlog.
  3. When interactions are crucial. Complex processes often require a wide range of interactions between people, tool and applications. Interactions, except when they cross the boundary, are difficult to identify unless you see them.
  4. When the output and the process don’t match. When, on occasion, the measured output or the output described by a manager does no match what is possible based on the published process then observing the real process is mandatory.

Once you have decided that you must observe, planning becomes a necessity.

  1. Begin by reviewing the known policies, culture and process of the organization or team being observed. This step helps to ensure that you have a sense of the environment and what you will be seeing.
  2. Decide on how long you will observe. Some processes and process variations need time to be seen. If a process requires a week to complete you will need to observe for at least that amount of time.
  3. Determine how you will record what you see. Trying memorize what you see will result in some information, however you will at least need to take notes. Remember that recording can include taking notes or recording audio and video. The level of detail needed will help determine the method needed.
  4. Finalize the logistics of the observation session before showing up. Office space, network and physical access can suck up huge quantities of time and effort. If you have a week for observation do not spend the first day dealing with administration tasks.
  5. Decide how you will create rapport with the group you are observing. Your presence will cause disruption. You need to find a way of observing with minimal impact to the results and without scaring those you are observing into calling placement firms. I am a fan of transparency; tell people why you observing and what will be done with the data. Where possible I usually involve those that I have observed in an early review of the data collected to elicit more information (hybridization of techniques by combining observing and asking).
  6. Finally do what was planned, but do not be afraid to tweak the plan as needed.

When I was in the tire plant, the time and motion guy would just appear and no one was thrilled. When we saw him coming we followed the proscribed process a bit more carefully, even if it was less effective. Observation can change behavior positively or negatively (Hawthorne effect). Sometimes observation might be the only way to know what is really happening, but without planning the data you gather might be what someone wants you to know rather than what you need to know.