skip to Main Content

DMAIC, which stands for Define, Measure, Analyze, Improve and Control, has provided a structure for process improvement for almost four decades. It’s an easy-to-follow five-step method that works in any industry and on any process.

Tune in to this 1-hour Introductory Webinar to get a primer on this how this handy model can help you in your quest to improve the world around you.

Webinar Level

  • Introductory


In this 1-hour Introductory Webinar, we will cover:

  • What is DMAIC?
  • Why do organizations use it?
  • How does it work?
  • What are some common missteps?
  • Are there any tip and tricks?

Webinar Transcript

Tracy O’Rourke: Hi, everyone. Welcome to’s webinar. Thanks for spending some quality time with us today. Hundreds of people have registered for this webinar and we’re really excited that you’re all here.

Lean and Six Sigma as you know are the go-to improvement methods used by leading organizations all over the world to delight customers, minimize costs, maximize profits, and develop better teams. Every month, we’re crafting webinars just for you, our global learner community. And our goal is to simplify concepts and tools of Lean and Six Sigma so that everyone can understand and apply them more easily and be more successful.

Today’s webinar is titled Introduction to DMAIC.

I’m Tracy O’Rourke. I’m a Managing Partner at And today’s presenter is also a Managing Partner, my colleague, the wonderfully talented, innovative, and consummately passionate about learning, Elisabeth Swan.

How are you today, Elisabeth?

Elisabeth Swan: I’m good, Tracy. Hello.

About Our Presenter

Tracy O’Rourke: Let me give you a little background on Elisabeth. She is the co-host of the Just-In-Time Café Podcast. She is a long-time Lean Six Sigma Consultant, Master Black Belt, coach, and trainer for over 25 years. And aside from almost three decades in the business, Elisabeth also performed with Improv Boston. And since half of process improvement is improvisation, that’s a great combo.

Welcome, Elisabeth.

Elisabeth Swan: Thank you again, Tracy. Good to be here.

How to Interact

Tracy O’Rourke: So before we get started, a few housekeeping notes. At the end of the presentation, we will have a question-and-answer session. However, please feel free to ask questions at any time by entering them into the question area. We welcome you to participate so we’ll also be asking you to vote in some polls.

And if we don’t answer all of your questions during the webinar, we’ll be sure to post the answers and share recording of this webinar and the slides on our website at

Where Are You From?

So please join us for our first activity, sharing where you are from. Hundreds of people have registered for this webinar all over the world and we just love to share where everyone is located. So please click on Ask a Question and type in where you’re joining us from today and we’ll give you a little time to respond.

All right. So I am seeing California, New York, New Jersey, Reno, the biggest little city. Thank you, Laura. Cleveland, Ohio, Kathy Rust, thank you for joining us. Raleigh, North Carolina, Columbus, Irvine, California, Helen at Montana, Charlotte from Las Vegas, Troy from Tampa, Florida, Minnesota.

Wow! We got a lot of people in the US joining today. We’ve also have someone from Saudi Arabia, thank you for joining us, and Toronto, and Stony Brook, New York, and the UK and Japan. Cody Bridges all the way from Japan.

Elisabeth Swan: Wow!

Tracy O’Rourke: He’s up. And Seal Beach, California. Patrick, I used to live there. That is super cool. All right. We got lots of people joining us from today, Elisabeth, to hear about Introduction to DMAIC.

Elisabeth Swan: That is great, Tracy. And I can’t see you guys but I’m psyched that you joined us. And Tracy, your introductions always make me feel better. So thanks for that.

Who Is

So as Tracy said earlier, we’ve both been with since its inception. And our mission is to make it easy for you to build your problem-solving muscles and that means we simplify complex concepts. We’ve made our training extremely practical and we think really enjoyable. We provide a running case study at the Bahama Bistro. Our restaurant team applies all the tools and all the trainings.

…our mission is to make it easy for you to build your problem-solving muscles and that means we simplify complex concepts.

And aside from this webinar series, we put out blogs, podcasts that Tracy mentioned, book reviews, lots of other information to help you get where you need to go. We’ve used and taught Lean Six Sigma for decades because it supplies the best toolkit for problem-solving.

We’ve Helped People From…

And thankfully, there’s a growing list of companies who agree with us. Here are some of the companies we’ve helped. As you can see, we’ve got brick and mortar, there are online companies, there are diverse industries like healthcare, financial services, manufacturing, and government.

And why did these companies use Lean Six Sigma? Well, because it’s about problem-solving. And once you have an organization, you’ve got problems. And like all of you, these companies want to be the best at problem-solving so you guys are in good company.

So more on some benefits later. But first, let’s go over today’s agenda.

Today’s Agenda

What is DMAIC? Let’s define what DMAIC consists of. Why do organizations use DMAIC? And how does it work? What are some common missteps when using DMAIC? And are there any tips and tricks?


Let’s dive right in and start with explaining what is DMAIC. So it’s this 5-step iterative method for improving existing process problems that have unknown causes.

The first step is Define where you define the problem. The second step is Measure where you quantify how bad is the problem. The third step is Analyze where you identify and study the cause or causes of the problem. And then the next step is Improve where you implement and verify the solution, verify that it works. And then the last step is Control where you maintain the solution. So you solve the problem, now you’re going to maintain the gains.

And it’s in a circle for a reason. I mentioned it as I described it, it’s called an iterative process. And that means you’re never done with process improvement. You can always do better. And you want that in a business. Things changed. Customers changed. Competitors changed. Technology changes. And continuous improvement is mindset. It’s a way of doing business. It’s a way of operating. And it is incredibly powerful to operate a business this way.

Now, let’s talk about what is this model and what is it not? It’s the model you use to figure out the root cause of an issue. It is not the model you use just to implement known solutions to known problems. So you’re looking for problems with unknown causes that need analysis, need some study to figure out what’s going on. The solution isn’t clear even though you probably have suspicions. You’re not really sure. Stakes are high, right? This isn’t how do we set up pizza Fridays better, faster? It’s a persistent problem.

Some people think that they have to put up with problems but this is the model you use to finally solve them and maybe pass solutions to solve this problem are ineffective and maybe you tried fixing it and you haven’t had luck. So that’s what it is. And that’s what it isn’t for.

Some people think that they have to put up with problems but this is the model you use to finally solve them and maybe pass solutions to solve this problem are ineffective and maybe you tried fixing it and you haven’t had luck.

Lean Six Sigma Project Types

So let’s take a look at what were the different types of projects? There are a bunch of them. And all of these, the DMAIC model is appropriate for the second one. That process improvement, that one in pink. But let’s look at what the other process types are.

Quick wins is the first one. These are great. If you have quick wins, that’s also a great thing to address. But they don’t require root cause analysis. That’s when you know the problem and the solution is obvious.

Now, the trick would be that people sometimes think they have a quick win but it actually requires some study. They don’t really know what the solution is. These are often called “Just-Do-It.” These are fast tracks.

And then you got process improvement. You can see at the bottom here, there’s the DMAIC model. They’re also called Lean projects or you can use PDCA known as Plan, Do, Check, Act or Plan, Do, Check, Adjust. We’ll cover that model briefly.

And then the next one is process design. It’s important to have a model to follow when there is no process. But DMAIC is for existing processes. You’re studying this as is process. And in the design process, there isn’t as is process. So you can’t study it. So that’s not going to work.

Process design is also the same, similar to process redesign. And that means there is an existing process but it’s so broken, you have to replace it. So that’s close to the design than DMAIC. And DMAIC is for iterative improvement, right? You just want to improve the existing, not wipe it out and replace it.

Lastly, there is infrastructure implementation, also really important. Organizations want to know how they’re performing. They want to know what their customers are saying, what their problems are. You need that kind of measurement and data collection. But it’s not an improvement project. There’s no root cause analysis necessary. So DMAIC doesn’t work for that either.

And people come up with projects and they will say, “Well, we need to implement this new software so my boss wants that to be my project.” Great. You should do that. But don’t waste time using DMAIC because it would not help you. If anything, it will just frustrate you.

So those are all the projects. The one we’re after is a process improvement project to work with DMAIC.

Poll #1

So now that you know about all of these project types, we want to find out which non-DMAIC projects have you been asked to use DMAIC on most? Because some of you, it’s already been involved in process improvement. And for you, this webinar is a refresher.

So for those of you in that position, I’m going launch this poll and see if you guys, what the experience is, those of you out there who have had some experience trying to use DMAIC. So go ahead and Tracy, you can – well, what you have seen, Tracy?

Tracy O’Rourke: Well, I’m going to go ahead and say that quick wins tends to be a big one is they’re sort of forcing DMAIC into something like training. Well, we need to train people. So how do I make that a DMAIC project? So not only do they have a solution, it’s typically a quick win.

Elisabeth Swan: Right.

Tracy O’Rourke: Or it might be some sort of software or even while there’s a process that exists, we’re not going to change it. We just want people to adhere to it. Well, that’s not DMAIC either.

Elisabeth Swan: Right. I’ve seen that. Yeah, just people – there is a good process. People aren’t following it. It’s like we don’t need a project for that either.

Tracy O’Rourke: Yes. So those are the ones that I see the most is people trying to apply DMAIC and not really going after the process analysis and root cause portion. It’s really more just a quick win. And I think part of it is because we all are sort of trained to jump to solution. And DMAIC doesn’t start with a solution.

Elisabeth Swan: That’s true. People have bosses or have corporate cultures where people, “Don’t bring me a problem, bring me a solution.” And that’s not going to be a good way to jump in to DMAIC either.

All right. Let’s take a look at what these guys came up with.

Tracy O’Rourke: So 42% was the highest, quick wins. So it sounds like people have also seen this happened. Process redesign 34%, followed by process design 18%, and infrastructure/implementation at 6%.

Elisabeth Swan: Interesting. I didn’t expect process redesign to come up at much but it makes sense. People know that this process is truly broken. They’re going to sort of rip it up and start again, which would be tough for this DMAIC.

Tracy O’Rourke: Yes. And I would also suspect that it’s a software solution, right? We’re going to redesign it. We’re going to implement software.

Elisabeth Swan: Yeah, yeah. And we’ve both seen that. Great. Thank you for that. Come back here. All right.


So, another important thing to look at is this other model, Plan, Do, Check, Act, because like DMAIC, it’s based on a scientific method and we’re going to actually cover that in the next webinar. It’s just a similar process improvement model.

DMAIC expands on a plan stage of Plan, Do, Check, Act. So that’s where Define, Measure, and Analyze can pair up against Plan. Improve is the do phase of PDCA. And then Control is Check, Act/Adjust. And it really focuses on this rapid improvement, rapidly refining and optimizing solutions in that phase. And it’s a good comparison to have as you move forward.

Using DMAIC to Improve a Process

Back to DMAIC. So when starting out and preparing to use DMAIC on a process problem, here are some classic challenges and best practices.

Finding good improvement projects can be difficult. You want to choose a project that’s focused on your day-to-day activities and a classic phrase we use with people is, “Fix what bugs you.” There are things that take you a long time. There are things you’re constantly having a redo. Those are great candidates for DMAIC projects because if you’ve been doing them for a long time that means you haven’t figured out how to solve that. Maybe you don’t even know why you have to do it.

You want to choose a project that’s focused on your day-to-day activities and a classic phrase we use with people is, “Fix what bugs you.” There are things that take you a long time. There are things you’re constantly having a redo.

The other big deal is to ensure leadership support. We always say, “If you don’t have a sponsor then there’s no project.” You’ve got to have support for what you do. People have to be willing to support you in your efforts to change a process.

Finding potential team members, also important unless people are familiar with the process. The greatest team members are process participants. And you want others to know about DMAIC, right? There are lots of – there’s no reason for other people no to be educated about this process as well. There’s free White Belt training, there’s free Yellow Belt training, and now there is this webinar. So everybody can get educated at some level.

Define Phase

The first step is the Define Phase. Once you’ve got a problem to focus on, you’re ready to dive into Define, getting this phrase right is like getting the trajectory of a rocket ship right. If it’s off, you might be heading to the wrong planet. But there’s an opportunity to adjust throughout the effort. This is where we check and make sure that everybody is clear on what’s happening. What are we deciding to do? What process are we working on? You’re always striving for clarity.

Define Phase Overview

So let’s look at an overview of the Define phase, four basic steps here. One, you’re going to create the charter. Create the charter means setting a goal like, “We’re going to reduce the time it takes to process an application from 4 days to less than 1.” That’s a great focus for a project. Put that right in the charter. Or, we’re going to reduce the amount of rework with applications from 20% down to 5%. Again, really great goal statement. It goes right in the charter.

And you have to be clear on the purpose. What’s the impact? What if you succeed in this project? Will the customers be happier? Will they be safer? Better able to do their jobs? Will you be better able to do your job?

Next step is interpret the voice of the customer. This means that you’re going to do interviews, surveys, whatever it takes to get clear on exactly what customers want.

I remember a long time ago working with a hospitality organization, the hotel company. And they were doing a pilot project to understand how customers felt about – how guests felt about the check-in time like what was a good check-in time?

So they were measuring how long it took when you are online. And they interviewed customers in the middle of this and they asked them just to get some color or get them to explain what did they expect like should it take 3 minutes, should it take 5 minutes, under 2 minutes? What was the expectation?

And they said, “Actually, as long as everyone is working and trying to see to every customer then we’re OK. We don’t like it when someone has their head down and they seem to be typing forever and they’re not looking at anybody. And we don’t like it when they go out that door, wherever that door is or wherever it goes to, that’s really annoying.”

So there were these other voices of the customer they found out from asking these questions. So you really want to talk to the customers of the process to find out, “Do we really know what they want? Let’s find out. Let’s talk directly to them.”

Next is understand current state. That means conducting process walks, mapping the process. It’s the best way to understand the process. There are lots of webinars and material about that.

And then develop project communication means figuring out who might be impacted by the project and speaking with them regularly. And that sounds easy but people get busy, they forget, and this makes life hard later on for the project. So we’ll come back to that to see what happens if you don’t regularly communicate with the folks that are impacted by process change.

Why a Define Phase?

So why do we need a define phase? Well, first off, you got to clarify the issue. When we coach people, we get people describing in their project charters things like, “My goal is to streamline the application process.” And I ask them, “What does that mean? What does the streamline mean? Are you going to reduce how long it takes? Are you going to reduce the percent of rework? What’s the focus?”

So it’s critical to gain clarity about exactly what you’re doing. If it stays fuzzy then you would not succeed.

So it’s critical to gain clarity about exactly what you’re doing. If it stays fuzzy then you would not succeed.

Another big question is scope, and this is another place where I find teams might think they are addressing application process but somehow, their manager thought they were fixing everything all the way to collecting payment. So clarity starts in define. You want to get that understanding and that clarity upfront.

Improvement Journey – Define Phase

And that brings us to the process improvement journey. So all of these phases are taking you on this journey, and the journeys are going to differ. And what you see here are samples of all the kinds of tools and concepts that help you along the way, either – a lot of these are images of templates or showing you what a process map might look like or stakeholder analysis or a model for how to work with teams.

So these are on the website. They’re all free. And you got lots of options. Let’s consider some define phase challenges and best practices.

One as I mentioned, you need a sponsor. You got to get the commitment from them and the team members. You got to draft your project charter with the team. That will help you build ownership. You can include them on the process walk, things like that.

You also don’t want to start with a solution. I knew that sounds obvious. But I recently got a goal statement that said, “By removing reviewers, we’re going to reduce approval time.” So people insert the improvement right in there. They pretty much come into a project knowing exactly what they want to do, which once again, I would say, “Well then, go do that. You don’t need DMAIC.”

Work with stakeholders. Involving the right people early is totally key. Publicize the process map. This is a great way to engage people.

Involving the right people early is totally key. Publicize the process map. This is a great way to engage people

We did some work up in Alberta, Alberta Health Systems. And we had – we called it a warmer. It was a huge room. We put a big map on the wall and we were trying to track exactly what happened from the moment a patient came in to admitting to the moment they were discharge. So it was a very big process, end to end process. We had lots of folks coming in to visit, doctors and nurses, and PTs, and OTs, and lots of people, everybody involved, the people from the lab.

And they would all clap and say, “Where did they experience problems?” They would point to it right on the map. It was a really great method of understanding. It also led to clarity. We had the admitting nurses. We had emergency room nurses finally talking to admitting nurses because they weren’t getting the right information and once they met and talked about it right there in front of the map, they realized it was one phone call which solves the problem and end this endless cycle of rework, of calling and checking the software to find patient information.

So the mapping is a great method for discovery right there. And it’s a great part of the journey.
So the point with that last example also is that everyone knew they were working to ensure that patients got good care. That was the purpose of reducing the cycle time from the admittance to discharge.

So back to that purpose, people always need to understand why are you doing this project, what’s the point.

Measure Phase

OK. So once the problems in process are clear, we get to the Measure Phase.

Measure Phase Overview

So the Measure Phase, first, we got to figure out what do we got to measure? If the project is to reduce application cycle time or something like that then cycle time is the measure and that’s why step 2 is plan for data collection. It seems obvious but sometimes I don’t even see them being measure on a data collection plan.

And what about separating that into different segments? What about measurement time it takes to reenter information? What about collecting data? What about the time it takes to call people to fill in the missing information? What about how long it takes to different people in the process to process the application? You can measure it by person. So, lots to consider before taking that last step which is collecting the baseline data.

Why a Measure Phase?

So why do we need this Measure Phase? Well number one, it’s to verify that there really is a problem. I’ve had a lot of teams who switched projects because well, they heard there was a problem or their boss said there was a problem, once they measured it, they discovered it was minor.

And in the process of proving that out, they find bigger problems to focus on. So getting data helps you build your understanding of the process and aside from that, you need a baseline. You’ll never know if you made a difference unless you know where you started. Again, that seems kind of intuitive where people get excited about making changes. So I’ve had team leads that started changing their process before they even got their baseline and then they had no way of knowing if they did any good.

Improvement Journey – Measure Phase

So those are some big reasons to have this Measure Phase which brings us back to our journey. There are things you’ll be completing as you go through the Measure Phase. There are templates like data collection plans and all these things will help you get the work done. Lots of images of the data you collect, different formats you could put it in.

Let’s talk about some challenges. So defining the measures. Does time to check in at the hotels start when the person got in line? Does it start when they get at their desk? And when does it end? When they got the key? What if the key doesn’t work? That happened to you, right?

Then they come back down. Is that still part of the check-in time? You haven’t got into your room yet. So translating what you not want to know into something observable and ambiguous is the goal. And there are lots of times that people haven’t been clear on that or maybe it’s tough term to clarify like what’s friendly mean? They have to measure that at hotels. How do they know if you’ve been friendly?

Well, one of the ways they know is when you call down and they use your name. They can look at the caller ID and they can say, “Yes, Ms. Swan? What can I do for you?” But it’s approximating friendliness. And they can measure if people use that technique or not.

Also, finding data is not always easy. You might think there are some great data sitting in a system somewhere. But you look at it and not everyone entered the data the same way or there are lots missing. Most of the time, you have to get it by hand at least to start and getting support and communication from others in order to capture data, that can also be tricky. People are busy. Include them in the plan. Let them design a data collection processes. Let them work on those with you.

People support things they have had a hand in creating. So let stakeholders know what and why you’re measuring. Some work environments don’t want problems exposed. The fear is it will make people look bad. So that’s an issue or it could be an issue. It depends on the culture.

People support things they have had a hand in creating. So let stakeholders know what and why you’re measuring.

Also, make sure people know the purpose of what you’re doing. What’s the result of reducing rework or cycle time? Why should they care?

And then will customers be happier, safer? Always remember to clarify the purpose of what you’re doing.

Poll #2

So based on some of those challenges and best practices, we want to get a little input from those of you that have tried this so far. So understanding that for some of you this is a primer but for some of you, this is a refresher.

So what is your most common issue in the Measure Phase?

  • There’s no existing baseline data? You can’t just go get the data.
  • There’s data but it’s inaccurate.
  • Our culture avoids problem data. Don’t want to see problems.
  • I need others to collect it and they’re not interested.

All right. So Tracy, I’m going to launch this. Let’s come over here. Go back to the next one. Launch that poll.

Tracy O’Rourke: It sounds good. And I would just say that I think number C is very interesting, our culture avoids problem data. And sometimes we know about it already that it could be a challenge and other times we find out after we’ve shared the data and the leader responded in a way that wasn’t something we want to promote.

So like, “Oh, I get it. The organization isn’t really game to share their problems with data.” And sometimes we find that out too late. And that’s why I always love doing this pilot DMAIC projects sometimes because it helps us see culturally how is this going to go in terms of finding issues and exposing them and making them more visible?

But I would have to say though, A is the one I see the most. There is no existing baseline data. And a lot of it is related to what you said earlier with operational definitions. People struggled with something they’ve never measured before.

Elisabeth Swan: Yeah.

Tracy O’Rourke: Friendly or even – we see a lot of cycle time measures but it’s usually the cycle time for a piece of the process and what we really need is the entire lead time from the customer perspective. And most people don’t have that readily available.

Elisabeth Swan: No. It generally leads to some form of manual data collection. And that’s what – people don’t expect that coming into these projects but it’s actually the reality is that you got to collect data yourself, which makes you understand the process better. It’s a good thing to go through.

All right. Let’s see what these guys came up with.

Tracy O’Rourke: OK. So it looks like same thing, 35% of the people said there’s no existing baseline data followed by B, there’s data but it’s inaccurate, which is unfortunate, and 21% say, “I need others to collect it and they’re not interested.” Yeah, that’s a cultural thing too. And last is C.

Elisabeth Swan: So that makes sense. The majority, there’s no existing baseline data. That was helpful. Thank you guys for chiming in. Let’s come back and move on to the analyze phase.

Analyze Phase

So once you’ve mapped the process and collected data, it’s time to study that map and study that data. So let’s come into the five phases in analyze process.

Analyze Phase Overview

So the first two steps are the analyze the process and the data. And that’s where you guys become detective. This is actually a great phase. This is kind of the heart of DMAIC. So you brainstorm root causes based on all your analyses. That means you need to include others.

So once again, including other people helps you get the collective wisdom about what causes process issues. People have different perspectives based on their job, their interaction with customers. You want to get that collective wisdom.

Then you develop root cause hypothesis and that means all your analysis has helped you to develop educated theories about what’s causing those issues. There’s usually more than one, more than one root cause. So validate root cause hypothesis means that you are not going by instinct. You need data. You need observation. You need interviews. Or some way to be sure, you’ve understood the problem correctly otherwise you might waste time in the improve phase and developing wrong solutions. So the goal is to unearth all the reasons for rework, the wastes, and all the barriers to flow, the why, the why in analyze phase.

Why an Analyze Phase?

I love our little doctor guy here. So think about what happens when you go to a doctor’s visit. You come in with symptoms, usually some kind of pain. And that’s often the result of an exam or a conversation with your doctor. We often get pain meds, pain medication. And we now have a national epidemic of over medicating for pain. Is that addressing the symptom or the root cause? Because they’re just addressing the symptom with pain, and that’s what happens in processes. We address the symptoms.

Examples like adding an inspection step to deal with defects. We think we’re going to prevent them from reaching the customers but we didn’t solve the problem. And the inspection doesn’t stop all the mistakes from reaching customers so we need to understand and solve the root cause. And that takes analysis.

Improvement Journey – Analyze Phase

So let’s look at this journey. Once again, lots of guidance, templates as you work your way through analyze. So let’s think about some of those best practices and challenges that you hit as you go along here.

So spending the time need to explore root causes. You can’t rush this. Good problem-solving takes time. So give yourself that time. You see a piece of a broken map up there on the right and then off to the left, you see some data, those are the ways you’re looking at your data, the ways you’re looking at your map and you can see a fishbone diagram. That’s the classic tool for brainstorming root causes.

One of the common mistakes people come up with are solutions that are really masquerading as problems. And the biggest one that Tracy and I see is a lack of training. The problem is generally people’s knowledge level. They don’t know the things they’re supposed to know. And training is a way to do that but it’s not the only way. So that could be – it might be just making an easier process. Why does it need so much training?

And I think about a great example of that is McDonald’s. Someone once referred McDonald’s to me as the largest organization in the world run by teenagers. Now, think about that for a second. It’s amazing that more people don’t suffer the ill effects of eating at McDonald’s.

But what they do count on is high turnover. They expect high turnover. They plan on it. They build their processes to handle it. So everything is visual. Their cash registers don’t have numbers. It just has pictures. So you don’t need to translate, “Well, let me look and see how much is a Big Mac. Now, let me type that or press that price in.” You just hit the image of a Big Mac.

So they have worked their process to be simple and visual as possible. They don’t increase – they didn’t think training was a problem. They just thought the process, understanding the process, the knowledge level people needed to have was the issue and they addressed it.

So the biggest issue in this phase is jumping to solutions without confirming the root cause hypothesis. This is expensive since there’s a good chance the assumption or the gut instinct is wrong and then it’s waste time and resources on a solution that may or may not address the true root cause. So spend time here to validate. Use the data. Use the observations. Use the interviews. Use the process maps, whatever it takes. This is the heart of what separates process improvement from just replacing old problems with new problems.

So the biggest issue in this phase is jumping to solutions without confirming the root cause hypothesis. This is expensive since there’s a good chance the assumption or the gut instinct is wrong and then it’s waste time and resources on a solution that may or may not address the true root cause.

All right. So that is the Analyze Phase.

Improve Phase

So let’s move on to the improve phase. This is the fun part. I think you got the impression by how many times that I’ve talked about or warned about jumping to solution that this is where everybody wants to be, right? And it’s back to that, “Don’t bring me a problem, bring me a solution.” We probably be just been trained over time to just be thinking about how to solve things. But now, we’ve done our homework, right? We define the problem. We measure the problem. We analyze it. Now, we’re ready. Now, we’re ready to look at how to improve the process.

Improve Phase Overview

So four steps here. So first off, crafting solutions means there are lots you can do. Involve one’s skin, the collective minds around you. Maybe you’ve been collecting great ideas all along. We always recommend people use a solution parking lot. You’re going to have great ideas from the moment you have that project idea and so is everybody else. So collect those because now, they’re going to come into play. And some of them might be obsolete now. But some of them might be great, might have some really great ideas. So let’s hear them. Let’s use them.

Then you got to filter as a solution because there are a lot of great ideas but you can’t do it all. And that might be just a case of we can’t afford it all. We can’t – maybe that’s one that’s timed resources. How much work does each of this take and what’s the return? How much is it going to impact my goal? How much is it going to reduce cycle time or reduce rework?

Next step is conduct risk management. You have decided which solutions you’re going to implement, now you want to make sure that this change you’re making or these changes aren’t going to come up and impact the process in a bad way. It’s like the ripple in the pond.

And a great example was I had a team lead working with a delivery team. She was trying to reduce the time it took to deliver packages. They had a fleet of trucks. This was at an oil refinery. And one of the issues they had to solve when they got down to what was getting in the way of some of these guys being able to make deliveries was somebody was or I’m not sure if it was one person or more than one person were siphoning gas out of the trucks that were left in the yard. So their solution was to install locked gas caps.

Now, they installed those but service slowed down even more and cost went up. So they had to dig into that and find out, well, what happened? Well, it turned out, people lost the keys to their gas caps. They had to drive to have the locks removed and then reinstall. So unintended consequences are one of t things you’re looking out for. And there are tools to help you think ahead and come up with those mistake-proofing ideas.

Another great example is you – some of you may, not without thinking about it, have these mistake-proofing around your house. Automatic garage doors, they’re fast, they’re convenient. They prevent your back stream, which I’m really was missing that during the latest power outage but since there is a risk, they might close on somebody. They came up with electric eye. In that way, they automatically stop if there’s a body or a bike or somebody is in the way. It will pop back up.

So you want mistake-proofing in processes. Great if it’s automatic. So it could just be manual. It doesn’t have to be involving technology.

And the last is, determine your solution approach. It’s time to pilot or do some rapid PDCA, Plan, Do, Check, Act cycles to try out your solution. It could be you solve the problem in phases. You got lots of options there. So those are the basic steps.

Why an Improve Phase?

And the reason you want an improve phase is, and this seems obvious, but did we actually solve for root cause? We have to check here. Did we reach out target? This phase might require multiple rounds to get to the goal to make it structured such that we check in the improve phase if we’ve narrowed the gap between the baseline where we were and where we want to be, which is our goal. Did we narrow that gap? And it might be a multi-faced gap but that’s OK.

So the analyze phase confirms the root cause. Improve phase addresses it.

Improvement Journey – Improve Phase

And let’s get back to our journey. So again, lots of helpful tools and templates at this stage so let’s look at some of the best practices and challenges of the Improve Phase.

So dealing with increased resistance when people realize that their area is going to undergo change, result of project is change. Somebody is going to have to change what they do. Some jobs are going to change. Some steps are going to change. And people might not like it and they might not even know what it’s about. So you have to involve the right people. You got to invite stakeholders to brainstorm solutions. It’s another opportunity for engagement. You always want that.

Somebody is going to have to change what they do. Some jobs are going to change. Some steps are going to change. And people might not like it and they might not even know what it’s about. So you have to involve the right people. You got to invite stakeholders to brainstorm solutions.

Just think about your own work experience. Have you ever gotten into your job and been handed a new form or been told about a new role? You don’t know where it came from, whose idea it was, whose decision. You’re just irritated because you had nothing to say about it. That’s how people feel.

I heard about a story from a large software networking company. This group was tracking down why some of them – they realized one of their root cause that they came up with is some of their components didn’t have serial numbers. So they couldn’t charge customers for maintenance if they couldn’t see that they were using their components. And it was about a $5 million problem. So, obviously important.

And when they told this department, this other department, that they had to start putting serial numbers on components, the response they got from these guys was I’d say it boils down to, “You’re not the boss of me. You can’t tell us that we’re going to change our job.” Well that – it got escalated. It didn’t happen fast. It didn’t happen nicely. Eventually, they got it through.

But if they had included this department in this process walks, if they had had communication earlier, they wouldn’t have gotten so much resistance.

Another example on a more successful side, this is a team from a child care non-profit. They were trying to reduce the cost of supply process. And this team lead, she got pushback on the changes she wanted to make. So one of the things she wanted to change was switch to a cheaper soap. And she got – people were worried about what is this new soap. Is this going to be an irritant? And she said she tried it out on herself. She let people know. She tried it. She used it. And she explained to everybody that it was used in NICUs, that’s Neonatal Intensive Care Units of hospitals.

And once she did all of that, people appreciated it and they said, “Great. Bring it on. Let’s save the money.”

So you have to do that essential reaching out. We force that at every phase but it’s a big deal on the improve phase. So lots of great tools here, lots of great templates and techniques, which brings us to the control phase.

Control Phase

So, as fun as the Improve Phase was, the control phase is often viewed as housekeeping or flossing your teeth. You have to do it. It’s not as fun but it’s actually setting the stage for more great improvements, and that is fun. Actually, improving stuff is a good time. I hear that all the time from people involved in this project, how much fun they had just solving this stuff.

Control Phase Overview

So let’s take a look at the four stages, the four steps in the control phase. First step is create the monitoring plan. No, you want to monitor the critical few measures. We’re not trying to do detective work anymore. We just need some critical few measures to make sure we keep an eye on what we improve.

We also want to come up with a trigger level. So if we got the delivery time from 4 days down to 1 day then if it goes over a day, maybe that’s the time to intervene. What’s the trigger point at which we intervene to see, “Hey, what’s going on here?” And that’s the response plan. You have to develop a response plan. So if you have to intervene, what do you do? So you want to think about that. Who do you call?

And then document the project so that others can learn from what you did. It helps with training new people, helps build momentum for more improvement.

And that brings us to the last step which is, pursue perfection. And that’s key. That’s the plan for continued improvement. Open the door for others to keep refining this process or replicate it. Where else can you use this solution? You just did a great thing. Can somebody else use that? So you want to keep getting the most out of your process improvement project.

Why a Control Phase?

Why do we have a Control Phase? The key is to not lose those gains. You did a lot of work. You put a lot in place. And if you’re not the process owner, you need to hand off the monitoring and response plan. You want to make sure somebody cares, somebody is watching. Make sure everyone is clear on the on-going process. You can’t drop the ball here. You’re building a culture of problem-solving. It’s key.

You did a lot of work. You put a lot in place. And if you’re not the process owner, you need to hand off the monitoring and response plan. You want to make sure somebody cares, somebody is watching.

Improvement Journey – Control Phase

So this last journey is both the end of our journey and then it’s also the launch point of your next journey. So again, you see great examples of templates and the work that happens in the phase. So let’s think about some of the challenges and best practices here.

So you want to make sure your improvements are solid before you’re off to the races to the next process to fix. Are there visuals? Are they easy to follow? Is it intuitive? Think about that McDonald’s. They make it so easy for everyone to do what they’re supposed to do.

And continue monitoring. New processes need care and attention. And the best way is to make monitoring visible and accessible. You can put numbers right on a whiteboard right next to the people who are doing the work. You can include process participants and the monitoring. You can do the hand-drawn charts. Hand-drawn charts and tables are fine.

And don’t forget. This process can still be improved and it should be. That same nonprofit child care client where the team lead was trying to reduce the cost of ordering supplies, once she announced that her savings enabled the organization to afford a whole new mortgage on a whole new day care center, all the staff got involved. She gets cost-saving ideas from her colleagues weekly. She has two suppliers, two external suppliers that not only decreased their prices, they became corporate sponsors of this organization. I get happy emails from her all the time. She is never done with this.

And don’t forget. This process can still be improved and it should be.

She’s doing another project but she loves that she can continue to update and improve this process. So there you go. It’s the restart of a new journey and the one you just fixed can always be made better.

Common DMAIC Missteps

Now, let’s think about some classic DMAIC missteps. And the DMAIC structure is great. But when it’s used for quick hits, it doesn’t work.

If you go solo, it’s all about engagement so you can’t and you shouldn’t take it all on your shoulders and you can help others build their skills by including them.

Jumping to solution, that means skipping the Analyze Phase but that’s the heart of DMAIC. Your gut instincts are right maybe 50% of the time, why risk it? Because you’ve done the rest of the work.

Lastly, assuming it’s done. Remember that team lead at the child care nonprofit? At the end of an effort, take a moment, appreciate the wins, let others know, celebrate. Find ways to expand.

In control phase, you’re going to monitor but it doesn’t the process can’t be improved even more. Improvements are not in concrete.

Question for You

So now, I’m going to reach out to you guys with the question. So for those of you that have had some process improvement experience, is there a misstep you’d like to add based on your experience with DMAIC?

So go ahead and answer that into the question box and once you answer those, Tracy is going to read those out so we can answer those.

But Tracy, of the ones I just mentioned, which one is your biggest pet peeve?

Tracy O’Rourke: My biggest pet peeve, well, I will say that jumping to solution probably happens the most often. But my biggest pet peeve is going solo, feeling like you’ve got all this knowledge and now you’re going to do process improvement to somebody as opposed to collaborating and bringing people with you.

But my biggest pet peeve is going solo, feeling like you’ve got all this knowledge and now you’re going to do process improvement to somebody as opposed to collaborating and bringing people with you.

So I think people, they get so caught up in the knowledge that they forget it’s a people thing too. I mean you want to bring people along. You want to – processes involve more than one person usually most of the time so that means don’t do process improvement to them. Bring them with you. Does that make sense?

Elisabeth Swan: Really good point. You make a great play on that. Don’t do it to somebody else. It’s something to do to a person. It’s to do with a person.

Tracy O’Rourke: Yes.

Elisabeth Swan: All right. What do you see?

Tracy O’Rourke: OK. So some of the other common missteps people are saying, lack of sponsorship or leadership commitment, right?

Elisabeth Swan: That’s a killer. There’s nowhere to go. Either – and it could be for lots of different reasons, maybe they said they were going to support and they didn’t but it basically means that project is dead in the water. You go to find that somebody will support and follow through on that support. I feel your pain. It’s not fun when that happens. But they’ve got to support it.

Tracy O’Rourke: Yeah. Kathy Rust wrote, “Going too big and losing sight of the original problem.”

Elisabeth Swan: Yeah, that’s a great point. That’s back to that clarity. It’s where DMAIC can really help as a structure to keep coming back to the charter. Are we still focusing on the same goal? Are we still focusing on the same outcome? Because you can stray. It’s amazing that it happens but you’re right. It gets bigger and bigger and bigger. It starts taking on more and more parts of the process. So holding that charter close and updating it is a great way to prevent that one.

Tracy O’Rourke: Laura Hall writes, “Analysis paralysis. Teams went off into non-scope areas because they are so busy analyzing.”

Elisabeth Swan: Yeah. It depends on the culture because I know especially in the hospitality industry, the hotel industry, they really want to just jump to it. But when I worked in the networking, the computer networking company, they were more prone to analysis paralysis. So you got to watch what’s your culture kind of support and encourage? So that would be something to watch out for when you got there because yeah, people can wallow in that and not get into Improve. Wanting to get it just right.

Tracy O’Rourke: Yup. And so, we have a few other comments from other people about like Kanda and Scott, both of them said, “Gathering too much data and then it’s difficult to filter.” And Scott had mentioned something about the data being compiled over a long period of time. It’s messy and daunting.

So, data issues were mentioned as well as no one owning the project. So, more issues about leadership as well.

Elisabeth Swan: Yeah. Those are great points. I’m actually – and that’s a great one you just brought up with data being messy because I’m about to dive in on some tips and tricks. So, that really tease up the tips and tricks. Let’s go there.

DMAIC Tips and Tricks

So one of the things you’ve got to do is reach out constantly. I mean the mention of the sponsorship and support of your projects, one of the people you’re reaching out to is that person in the leadership position who said that they would sponsor you. They’ve got to know what you’re doing, what did you find.

Often, they have helpful comments around which direction you should go. You may come to, “There are two big problems in this process. Which one should we go after?” Because they’re going to require different analysis and different solutions, so keep them really close to the project as well as all the stakeholders.

The next one is structure sets you free. So DMAIC provides guidance but you might have a slightly different path. You don’t have to – there’s not a by-the-book way to go through this. And I think about this great example. This is a different organization I was working with recently. The team lead was in charge again of a courier group and he used the spaghetti map.

…structure sets you free.

Usually, you use a spaghetti map, it’s kind of like a way to trace all the movements in a workplace. You could actually think about people walking around and seeing the outline of their steps as a form of spaghetti.

But he used it to draw the map of where were they driving to get to their destinations and then compared it with Google. So he went off on his own on how to use the tool and that was just a great example to me of not being – feel like you’re held in this straight jacket of what DMAIC does. So it gives you structure but you can move within that structure.

The next one is back to the listener who just said data is messy, and that is so true. It is not sitting there for you. You have to go get it. It’s not in the form you want. It might not be accurate. You have to dig into the data. You have to filter it. You might want to get the most recent. You have to be very clear of what you need and what you don’t need. So that is just a reality of data. So you really got to work that side of it to make sure you can make it useful for you.

The other one is analysis has many forms. You could observe the process. What are people doing in the process? You can study the map. You can talk to people. You can analyze the data. So there are lots of ways to analyze and understand the process.

Another one, I think this gets back to that listener’s comment about analysis paralysis. And this – I love this phrase, “Don’t let perfect be the enemy of good.” Implement your idea. Don’t wait until you’ve designed it perfectly to the nth degree. Try it out. Tweak it. DMAIC is iterative, right? Do it again. Do it again. Get improvements. Get it working. And then get the feedback. Don’t wait.

And lastly, go immediately to your next project. Do not let your problem-solving muscles atrophy. It should be just the way you do business. You should always be looking for problems to solve.

Do not let your problem-solving muscles atrophy. It should be just the way you do business. You should always be looking for problems to solve.

Today We Covered

And that brings us to our agenda. We covered what DMAIC was, why organizations use it, how it works, the common missteps, and the tips and tricks to get around it.


So what we’d like now is for other questions you guys might have. We want to hear from you.

Getting Started

And while you think about entering your questions in the question box, we’re going to go over a few things that will be helpful for you guys to know. You can learn more about DMAIC by signing up for any of these trainings and certifications. This might be the start of your journey into Lean Six Sigma and it’s a big world so lots of tools and templates for you to learn about in the trainings.

And this is the bonus for having joined us today. You guys get 20% off before the end of this month. You just use the code 20DMAIC listed right there.

Upcoming Webinar: April 20, 10am PDT

And then there’s another webinar come up which I mentioned. Tracy, do want to give these guys an idea of the next webinar?

Tracy O’Rourke: Yes. So it’s Introduction to PDCA, which is Plan, Do, Check, Act/Adjust. I mean it’s really going to be very similar to the format that Elisabeth did today on DMAIC. They’re surprisingly similar and we’ll just talk about the differences between the two as well as the steps PDCA.

Just-In-Time Café Podcast

Elisabeth Swan: And the next podcast – so the podcast is out right now. Tracy, do you want to talk about your interview with Karen Martin?

Tracy O’Rourke: Sure. So Karen Martin, if you haven’t heard about her yet, I’m shocked and it’s because she has written five books. She is a very popular in Lean and the Lean community, building lean. She is very accessible on LinkedIn. She has got free webinars. And we talked about her latest book Clarity First in our interview, and it was really awesome talking with her, she is very enjoyable and entertaining as well as smart.

Success Story Webinar

Elisabeth Swan: So that’s a great one. Let’s check in to our podcast if you haven’t already. And there also is the latest success webinar. Tracy, do you want to tell us about?

Tracy O’Rourke: Sure. So this is – success stories are basically successful Green Belt projects for the most part but not always Green Belt. Sometimes they are black belt projects or sometimes they’re just reducing waste.

And so, this is really around application, how successful two people were or one person is with their project. And in this particular case, these two girls, Kymberly and Darlene work at UC San Diego and they did a process improvement for students who have to submit an event. They have to submit and get the event approved before they can have it. And they went from 27 steps to 7 steps, reducing lead time by 81%. It’s a great success story.

Elisabeth Swan: And those are all accessible for you guys if you want to see examples of successful projects.

Wonder Women of Quality

And last stop, we’ve got a new series called the Wonder Woman of Quality. It’s like the Me Too Movement for quality because usually, it’s men of quality. So Tracy, do you want to tell us a little bit about Eunjoo?

Tracy O’Rourke: Yes. So I worked with Eunjoo for a very long time. And she is – I’m very happy and proud that she got nominated for Wonder Woman of Quality. She has done an amazing job in King County in the financial group and she has done some amazing work with leading an entire organization and promoting leader standard work within their organization. And it’s great. You should check it out. She has done a lot of good work. She is an inspiration for many.

Elisabeth Swan: She is.


OK. Let’s come back. We’ve got a little time and if we don’t answer the questions on this webinar, on the next three minutes, we will post the answers to all your questions online.

Tracy O’Rourke: Right. So we don’t have very many questions for this. But we do have two from Anush so I’ll go ahead and they’re sort of similar so I’ll go ahead and give you that question.

And the question is around baseline data. So he first writes, “What do you mean by baseline data in the Measure Phase?” And I’m going to just give you a buzz so you can think about it. And so basically, what do we mean by that? And if you have a bunch of it, a bunch of data, how do you find the baseline data?

Elisabeth Swan: That’s a great question. And there’s actually two webinars, one of them is about collecting data. Another one is about displaying your data. So you should definitely look those up for more depth on this.

But your baseline is basically the most recent data around your main matrix. So if you said that you’re going to reduce the amount of rework then you want to measure how much rework is there in the process in the process now? And that might mean you can look and see how often applications were sent back to be fixed or you might have to collect it live and ask people every time they get an application and there’s wrong information or missing information, you want them to write that down. And then you take the total number of applications for a month and you look at all the times that people had wrong or missing information and you get a percentage. That’s your baseline.

So it depends on what your matrix is. It depends on whether you have existing data or you have to create it yourself and you have to get enough information to feel like, OK, now I understand how are we doing on cycle time right now. How fast does it take us to check in guest based on the average of the past three months? Or how much rework is there based on the data we just collected for the past month?

Hopefully, that’s helpful. But I think there are definitely some resources for you if you want to do exploration of that. And of course, there’s a free Yellow Belt Training. You can go to check out the measure phase there.


Tracy O’Rourke: So Elisabeth, unfortunately, that is really the only other questions that we’ve gotten but I will say that someone said, “Thank you. This was great for my Green Belt Training.”

Elisabeth Swan: Well, that’s awesome. You’re welcome. Thanks for joining us, everybody. It has been a pleasure having you.

Past Webinars

View our upcoming webinars and join live so you can ask questions and let us know what you’d like to us to cover next. We’re busy building new webinars all the time. And we’re happy to know you’re busy too – building your problem-solving muscles – keep it up!

Get Full Lean Six Sigma Training & Certification

You can also register for our full Yellow Belt, Green Belt, Black Belt and Lean Training & Certification courses to give you a solid foundation for applying Lean Six Sigma.

Elisabeth Swan

Elisabeth is a Master Black Belt at, the co-author of The Problem-Solver’s Toolkit and co-host of the Just-in-Time Cafe. For over 30 years, she's helped leading organizations like Amazon, Charles Schwab and Marriott International, Inc. build problem-solving muscles with Lean Six Sigma to achieve their goals.

This Post Has 0 Comments

Leave a Reply

Close search
×Close search