Locked lesson.

Upgrade

About this lesson

The X/Y Matrix is a tool that maps the Voice of Customer needs and concerns (Y) onto the process steps and inputs (X). This tool can take on several different formats, but any of them will provide the Lean Six Sigma team with insight into the contributing factors of customer value.

Exercise files

Download this lesson’s related exercise files.

XY Matrix Exercise.docx
61.6 KB XY Matrix Exercise Solution.docx
61 KB

Quick reference

X/Y Matrix

The X-Y Matrix is a tool that maps the voice of customer needs and concerns onto the process steps and inputs.  This tool can take on several different formats, but any of them will provide the Lean Six Sigma team with insight as to the contributing factors of customer value.

When to use

The X-Y Matrix is used to prioritize the areas of focus by the Lean Six Sigma teams during the Measure phase.  If there are many possible areas of focus, this is an excellent tool.  If the area of focus is clear, the matrix is not required.

Instructions

The X-Y Matrix is used to relate customer needs to process inputs and steps.  There are many different forms, formats, and scoring rubrics, but they all do essentially the same thing.   They relate the process inputs and steps to the customer needs in order to prioritize the areas of focus for the measurement and analysis.

One side of the matrix (usually horizontal) has each of the customer needs or project “Y” goals. These are often weighted based on customer input with the weighting from 1 to 10 with 10 as the most important. The other side of the matrix lists the possible “x” inputs to the problem or the process steps. Each “x” factor is then related to the “Y” factor usually with a relationship of low, medium, or high. These ratings have a rating of either 1, 3, 9 or 1, 5, 9. Each cell of the matrix is assigned a score based on the relationship rating times the weighting factor. All the cells for a given X factor are summed to create a total for each factor. These totals provide a relative importance of each factor and should guide the data collection process.

When evaluating the results of an X-Y matrix, the columns with the highest total are the columns that have the most significant impact on meeting the customer’s needs.  These inputs/steps need to be well-designed and well-controlled.  If a column total is zero, that is a non-value-added input/step and should be considered for elimination.  If there is a row that has no interactions, that is an unmet customer need and the process should be redesigned to address it.  Again, the score for each column does not indicate if that step is operating well.  The step may be excellent with no room for improvement.

One of the more commonly used techniques for X-Y matrix is the Quality Function Deployment or QFD,  The QFD, also known as House of Quality, is a very robust tool and is excellent for developing new products or services,  Within Lean Six Sigma, I just use the most basic matrix within, QFD, which is an X-Y matrix.  Note that the QFD format switches the rows and columns as compared to most X-Y matrix formats.

To create a basic QFD:

  1. List the customer needs from the Voice of the Customer research.  If possible, weigh those needs using a 1-5 scale, with 5 being the most important and 1 as the least important.  You can have multiple needs with the same weighting score.  This list is on the vertical side of the matrix.  Each customer need is a row in the matrix.
  2. List the process inputs and process steps across the horizontal side of the matrix.  Each input and step is a column in the matrix.
  3. Classify the strength of the relationship between the input/step and the customer need using High, Moderate, Low, or none (represented by a blank cell in the matrix).   When determining the relationship, ask the question, “How important is the input/step to the process’s ability to meet the customer's need?”  Note: you are not evaluating how well the input or step works, only strong is the relationship between them.
  4. Score each relationship by setting High = 9, Moderate = 3, and Low =1, then multiply the relationship value by the weighting value for that customer need to determine the value of each cell in the matrix.  Sum the columns to determine the overall score for each input/step.

Hints & tips

  • Don’t overly agonize over the ratings, If you are off on a few, it probably will not change the relative rankings of the columns.
  • The actual factor or step value is not important, it is the relative value as compared to other factors or steps.  Each process will be different because of different customer needs and the weighting of those needs.
  • If your organization uses an X-Y matrix with a different scoring system – go with it.  I have seen H/M/L scoring of 9/3/1, 10/5/1, 3/2/1, 7,3,1, and others I don’t remember.  Regardless of the scoring technique, the relative column ranking was essentially the same.
  • If your customers will not give you weighting, then just set the weighting value for all customer needs at 1. 
  • Be wary of letting people in the process set the customer need weighting.  They have a tendency to weigh things based on what they find most difficult, not what is most important.
Login to download
  • 00:05 Hi, I'm Ray Sheen, working with a large complex process or
  • 00:09 one with many inputs and controls, you want to focus on the critical ones,
  • 00:15 the X-Y matrix is a tool that can help.
  • 00:17 >> We've already talked about one prioritization tool, the Pareto Principle.
  • 00:22 This approach is a little more subjective, but
  • 00:24 provides excellent insight because of the multi-dimensional aspect.
  • 00:28 Let's take a look.
  • 00:29 The X-Y Matrix is a graphical representation of the process.
  • 00:33 In this representation, we can see how inputs and
  • 00:37 controls are related to the process outputs.
  • 00:40 We can then use this to prioritize which inputs or
  • 00:44 process steps we should focus on in the data collection and later in the analysis.
  • 00:49 Now, there are many different formats and
  • 00:52 scoring approaches that use the X-Y Matrix.
  • 00:55 From what I've seen, all of them will work.
  • 00:58 Some take a little more time, but provide greater insight.
  • 01:01 Some are easier to score, but are a little more subjective.
  • 01:05 At the end of the day, what matters is that you use a format that works with
  • 01:09 your team and that provides helpful insight and
  • 01:12 prioritizing the areas of data collection and improvement.
  • 01:16 One point though, about the different approaches,
  • 01:19 all of them use voice of the customer for inputs.
  • 01:22 And If there is a common problem that I see,
  • 01:25 it's that teams don't really use the voice of the customer.
  • 01:28 They assume they know what the customer wants and thinks, so
  • 01:32 they use their own interpretation of what they feel the customer should want.
  • 01:36 I have found that when we use an internal team definition of the voice of
  • 01:41 the customer, we are almost always wrong.
  • 01:44 So get the real customer and get their input.
  • 01:48 Let's take a look at an example of an X-Y Matrix.
  • 01:51 A common application of the technique is to use it as a means of prioritizing cause
  • 01:56 and effect relationships.
  • 01:58 In this case, process inputs can be mapped onto the CTQs,
  • 02:01 which are the process outputs, that we are most concerned about.
  • 02:05 In addition to mapping the inputs to the CTQs,
  • 02:08 we could map them onto process steps or even SIPOC categories.
  • 02:13 The matrix is set up so as to allow for
  • 02:16 the prioritization of the X attributes in the Y=f(x) equation.
  • 02:22 If there are multiple Y's, each should be a separate column in the matrix.
  • 02:27 A frequent element in the analysis is to weight the Y factors to those
  • 02:32 which are most important.
  • 02:34 Now, a common scale is 1 to 10, where 10 is the highest priority or
  • 02:38 the highest importance.
  • 02:39 And 1 is used with very low priority or low importance goals.
  • 02:45 Next is to decide on the likely relationship between the X factor and
  • 02:50 the Y result.
  • 02:51 I usually use a 1-3-9 scale or occasionally a 1-5-9.
  • 02:56 If the X has very little effect on the Y result then it gets it's a 1.
  • 03:00 If it's a large impact, we give it a 9.
  • 03:03 If it's somewhere in the middle, we'll either do the 3 or the 5,
  • 03:07 depending on the scale I'm using.
  • 03:09 To get the rating, multiply each of the relationship scores by the weight value,
  • 03:14 then add the scores across the row.
  • 03:16 The totals you get will tell you which factors appear to be the most significant.
  • 03:21 One of the more prominent techniques that uses an X-Y Matrix format is
  • 03:25 the Quality Function Deployment, or QFD.
  • 03:28 QFD is an advanced version of the X-Y Matrix that applies the relationship
  • 03:33 analysis along multiple dimensions.
  • 03:36 It's sometimes called the House of Quality.
  • 03:38 When you look at the diagram,
  • 03:40 you can see that it looks kind of housey with a peaked roof and a side porch.
  • 03:45 The QFD relates our customer needs or wants, those are listed on the side porch.
  • 03:51 To the product features or the process steps, those are listed across the top.
  • 03:56 Now, I've used a QFD with both product and process analysis.
  • 04:00 In the diagram I'm just showing a dot where the relationships occur.
  • 04:03 When I do the actual analysis, those will be a one three or
  • 04:08 nine based upon the relationship between the customer needs and the process steps.
  • 04:14 The QFD can provide additional analysis beyond just the basic
  • 04:18 customer needs to process steps.
  • 04:21 We can analyze each of the process steps by breaking those down and determining
  • 04:26 which of those steps have the most impact on the overall performance of the process.
  • 04:31 We can do several iterations of QFD, each of them being an additional X-Y Matrix.
  • 04:37 >> The X-Y Matrix is a great way to prioritize your data collection and
  • 04:41 it can point to some likely areas for improvement in your process.

Lesson notes are only available for subscribers.

PMI, PMP, CAPM and PMBOK are registered marks of the Project Management Institute, Inc.

Gift this course