Attribute Results to Processes
This may be the most controversial point, but in my opinion it is aligned with the r
est of Deming’s philosophy nicely, and I agree with this point totally. In the US especially, Management By Objectives (MBO) is very much the status quo. I’ll give a short explanation of my opinion from an operational standpoint first before relating this concept to project management.
In MBO, standards are set for a particular process with the intention of evaluating employee performance by them. Performance in relation to the standard weighs heavily (and is sometimes the only factor) in merit increases, bonuses, etc. A standard example would be handle times in a contact center environment. If the standard handle time is set to 3 minutes, and you are taking an average of 4 minutes, it is said the employee is performing poorly. This paradigm assumes that the measured metric (handle time) is the full responsibility of the employee. Pros and cons of this management philosophy:
Pros
- Easy to set expectations
- Easy to quantify
- Easy to base performance evaluations on
Cons
- Tends to move the focus to being ‘at’ standard
- No focus on process variability
- Tends to make the standard the ‘only’ performance measure
- Provides little incentive to improve processes when the standard is being met
- When defining standards, operational leaders tend to lean towards lower standards in fear of not meeting them
- Propogates the ‘carrot and stick’ approach where fear usually wins out as the strongest motivator
- Discourages educated risk-taking and experimentation with processes, because it might throw people ‘out of standard’ temporarily
- Discourages employees from helping each other by encouraging detrimental internal competition
- Forces employees to try and acheive contradictory goals (”I want to provide great customer service and spend the time the customer requires, but if I do that the way I know it really should be done, I’ll miss my standard and get written up.”)
- Employees and managers may be motivated to skew results so their numbers look better
- Moves focus from customer satisfaction to “covering my butt.”
You can probably tell that I don’t like Management by Objectives. To me it seems like the easy way out, and very much the wrong approach. Deming would say that 90% of defects in any situation can be related to poor systems or the lack of systems in place. Most people want to do a good job and will follow a process when it is well designed and they have the ability to provide input for it’s development and improvement.
Let us discuss this point in terms of estimating project tasks for duration and cost. In the MBO paradigm, what usually happens is that a project team member is given a piece to plan and estimate. Many times there is no process for them to follow in making their estimates. The project manager assumes they are the SME and know how to estimate. The PM may not really have a good idea of how much time they will be able to devote to their project work, alongside all their other projects or daily duties. In many cases an experienced team member is going to throw on a lot of slack time because they are in fear of missing their estimated deadlines. In any case, an MBO mindset is going to lead everyone to blame individuals for mistakes. It doesn’t necessitate a focus on improvement.
What would a Deming approach change? First, there would be a process in place to help guide estimates, evaluate performance to planned estimates, and go back to figure out why estimates were wrong. Another option might be to change the resource load so that a team member can devote all or most of their time to a project for a limited period of time, thereby reducing the cycle time on their deliverables (for your and other projects) and allowing them to more easily estimate in terms of effort required. Part of the process may be to train and guide them in doing a lower level of WBS to break things down into 4-16 hour chunks. It is going to be important in a Deming approach to evaluate tasks that took longer or shorter than anticipated. Not to place blame on the individual who did the estimate, but to find ways to enhance the process of estimating to make it more accurate in the future.
(Back to Deming's 14 Points)
