Wednesday, May 27, 2015

Which Matters More? Process or Results?

If the room is too cold, do you work on the thermometer or the heater?

Which is more important . . . process or results? It's a question that spurs debate on bulletin boards and in blog posts. The problem with the question is that it approaches the two as disconnected rather than one-in-the same. Obviously, results are what really matters, processes exist to produce results, and when results are not at the levels desired, the processes need to change. On the other hand, whatever results are at any given time, it is impossible to sustain them successfully without effective and stable processes.
Results Drive Process . . . Which Drives Results
A manager who emphasizes results over process will drive team members to deliver "at all costs." The range of possible outcomes from this type of a culture include continual quality and safety problems, regular cost overruns, and high employee turnover.
On the other hand, a manager who understands the Plan-Do-Study-Act (PDSA) cycle drives and supports team members to improve processes based on the results they are getting. The outcome from this type of focus is continually improving processes and a much better chance of sustaining results.
Those who argue against a process focus often cite examples of long, complicated instructions that are difficult to follow and cause more problems than they prevent. This type of process documentation is not, however, aligned with a process-focused mindset. The instructions are often written by people who are not doing the work and do not exemplify PDSA thinking. When problems arise, the instructions are either not changed quickly to address the root causes or the changes are additive – resulting in longer and more complicated documents.
The essence of PDSA thinking is assuring processes are achieving desired results on a continuing basis and, when they are not, improving them. The key word being improve. Changes to processes should include making the process easier and simpler to meet standards and produce desired results. Changing a process by making instructions more complicated and difficult to follow does not constitute improvement. A new or revised instruction that is difficult to follow should identify a problem that drives further change – and the change should be to simplify the instruction.
When an organization without a process focus misses its targets, improvement will be futile because there is nothing with which to maintain the improvement. When a person or team finds a better way to perform the process, there is nothing to assure that a different person or team will follow the new way.
Letting a process run without standardization is damaging to the culture, teamwork, and customer. Letting people do things their own way also goes beyond the shop floor and bleeds into hiring, leadership, and a host of other aspects of the organization.
Leading and Lagging Indicators
Lagging indicators are the measures of the results of a process. Leading indicators are the upstream measures that enable problems in to be seen before results are affected. The key to developing effective leading metrics is to make them activity based, meaning that they measure an activity the team performs that can be quickly adjusted when a problem arises. Along this line of thinking, effective leading metrics often measure adherence to critical parts of the process. When a process focus is absent, however, this type of measure will identify a lack of standardization and stability but, without a change to the company’s culture, will not prevent problems or drive improvement. Lagging measures – or the results – will continue to be highly variable, putting further pressure on results.
Revising the Question
Results always matter. When one understands the PDSA cycle, however, it is not a question or process versus results; it is more a question of whether processes matter or not. In other words, working on the thermometer will not warm up the room but focusing on the heater, insulation, etc. will.

No comments: