If you’re like me, you’ve been seeing, saying, and maybe practicing the method behind the acronym PDCA throughout your entire Continuous Improvement career. Ever done a root cause analysis? A kaizen event? A DMAIC project? Or pretty much any Lean tool for that matter? Yup, you’re in this too! You were probably sucked into a PDCA without even knowing that’s the kind of party you were going to. Either way, there’s a good chance you’re doing it all over the place; and there’s also a damn good chance you could do it bigger, smaller, and just better in general.
What is PDCA?
“Plan-Do-Check-Act”, “Plan-Do-Check-Adjust” or “Please Don’t Call Again”, I think you can go either way on this one. If you bust this method down to its core elements, it’s really just the good ol’ fashion scientific method (it’s still okay to use the word scientific, right? Just checking). Let’s do a quick and dirty side-by-side:
Plan – Form a Hypothesis
Do – Run an Experiment
Check – Validate if the hypothesis is true or false
Act / Adjust – Decide whether or a new experiment is needed and repeat
Yes, I know this is oversimplified and there’s a lot more to running valid experiments and all that – but you see what I’m getting at. Now we have a pretty good idea how this works on a project level. Let’s say we want to fix an issue with a failing motor that’s causing excessive downtime. We may go out and observe (plan), troubleshoot, try something (do) – oops that didn’t work (check) – then try something else (act). Then keep trying until we figure it out and just continue to monitor until we’re comfortable that the dang thing is fixed. This is PDCA in it’s simplest form.
Let’s scale up one level to the kaizen event. Now we have a system of conveyors and machines that is dropping $250k / year worth of product on the floor from various places. We might capture the waste at different points to measure, then quantify losses per leak. Then we might prioritize which leaks to go after and in what order (plan). Then start the root cause analysis process or “where-where-why-why” analysis and then start troubleshooting (do). From there we continue to collect data to validate that losses have in fact been reduced (check). Finally, after learning that losses have actually gone up, we fire the Continuous Improvement leader, reassign someone who knows what they’re doing, and try again (act). Okay, let’s keep the CI guy and give them one more chance but they’re only allowed to 5S the front office.
Now let’s zoom out. Past all the engineers and accountants. Past the 17 levels of middle management. Past the 900 Vice Presidents and baam!…you’re the CEO. Your stock values are down. Wall Street is sending you death threats if you don’t do something quick. You hire a top tier Management Consulting firm to come up with a brilliant strategy (plan). You deploy this strategy throughout the company sparing no beating heart in the company (do). After about 6 months, you patiently watch the stock value barely budge (check). Then you decide the strategy and the consulting firm were bogus, you scrap them both and go with your gut to transform the company (act). Yes…this is also PDCA.
Where PDCA Fails?
Well since there are four letters involved, you guessed it, there are four places where PDCA can fail. Let’s just go in order, shall we?
Plan – You’ve heard the expression – failure to plan is planning to fail. Planning indicates the intent to prosper from an activity or at least avoid unnecessary losses. A good plan should define parameters around who, what, when, where, why, and how. It should also clearly state the hypothesis and how it will be validated. If controls can be put in place to make sure other “issues” don’t skew the result, that’s even better. It’s also good to establish from the onset what “success” looks like and the reaction protocol when (not if) it doesn’t come out as planned the first time around.
Do – This one goes without say. A wonderful plan that never gets executed is nothing more than a wish. You may think this is obvious but consider that 90% of strategies never get deployed. In fact, a whole world of executives will cite poor execution as the biggest reason for failure in business overall.
Check – Believe it or not, many-a-kaizen event are done and wrapped up with no performance tracking mechanism in place to validate if the dang thing even worked. I mean, people will go all out on the celebration after a kaizen event – popping champagne bottles, handing out t-shirts, big smiles for pictures in the newsletter and all that. But if you go back 3 months later and ask if the results have sustained, and you get a bunch of blank stares. Of course this has never happened to you, but take my word for it that it happens.
Act – This is a good barometer for the true Continuous Improvement mindset and culture at your company. You have to have a growth mindset to be any good at this one. You have to be willing and able to accept that perhaps your hypothesis was wrong and that the first go round was a learning exercise. Then be willing to give it another go based on your new knowledge to maybe get it right the next time. And then be willing to do this over and over and over until you get the result you want. If you have a succeed vs failure mindset (as opposed to a growth mindset), you might just write the whole thing off as a failure way too soon and accept the current state as a fact of life. This failure mindset is fatal in Continuous Improvement and should be discouraged at all costs. Do it for the culture. Do it for … the people!
How to get Better PDCA’s?
Better PDCA’s, by definition, comes with improved quality and increased quantity. You have to be deliberate about making sure each step is executed the right way, every time it’s supposed to be. I watch my 7 year-old son execute PDCA over 100 times to beat a level in Mario Bros. Yes it took him 100 tries but he got it…and for him, it was totally worth it.
Here’s the gameplan:
1) Start with the company’s strategy so everyone is clear on what’s most important to improve
2) Then challenge every employee to improve something important in their area of ownership
3) Use Lean tools such as Root Cause Analysis, Kaizen, 5S, and others to identify needed changes and make them
4) Monitor the results and adjust / respond as needed. It’s best to have strong analytics system in place so tracking data doesn’t seem like added work
The key to success is not quitting until you get it right. In fact, the definition of success is not quitting (paraphrased for your convenience). This works at all levels in the organization so no one is immune. Implement a management system like impruver.com to facilitate this practice and develop this capability throughout the entire company and see a dramatic and permanent shift in culture, capability, and business results.