Archive for June, 2009
Find a process, Find Quality
Posted by: | CommentsQuality in marketing may be considered too subjective and therefore difficult to define. However, there is a lot about quality and marketing that can be made objective. But to first understand quality in marketing, I think it is important to understand two terms: Process and Metrics.
Quality is not an event – it is a process, and a mindset. A consistently high quality product cannot be produced by a faulty process. There needs to be a repetitive cycle of measuring quality, updating processes, measuring, updating processes, etc. In Six Sigma we use a process called DMAIC which means Define, Measurement, Analyze, Implement and Control. In Lean they discuss the PDCA cycle of Plan, Do, Check and Act. You cannot have a quality system without a defined process. So we must start here and using an established process is simply the best avenue that most of us can take.
To make the quality process work, collecting metrics is vital. If you are not going to capture metrics, it will be hard to improve processes through a quality management initiative. So, what kind of metrics do you capture? First, we must break down the term of “quality” into a number of areas that define the characteristics of quality. Then, look at each of the individual characteristics and determine one or more metrics that can be collected to mirror the characteristic. For instance, one of the features of a quality e-mail program may be increasing percentage of open rates. This is an easily achieved metric that can be measured after the solution goes live.
Releated Subjects: 10-Step Project Management , Lean Six Sigma
If you control it well, it flows well!
Posted by: | CommentsSimple, concise control is now what you need. If we cannot build it into the project we have to go back to the Measure phase or even to the Define phase. Without controls, we will return to the old way of doing things and when asked about it three months later we will say, “Oh, we tried that, it just did not work!” If you did your job, in the previous stages, the controls that are needed should be readily apparent.
Control is about having a documented repeatable process that can be monitored for variation. Walk the participating parties through a mock exercise and see what takes place. If you want, for the fun of it, videotape the process. Are you ready?
My Control process basically consists of:
1. The operation is documented.
2. Process control plans are created
3. Scorecard is created
4. Problem encounter plan
You just did your walk through and created your operation plan. Just write down what you are going to accomplish and how you are going accomplish it. Pass it around and get an agreement on it, not only from the people that developed it but the people that are performing the work. Check once again if it satisfies the customers needs. After that, draw a concise process plan. This is your final checklist on how you are going to go about the implementation. If you are not satisfied, you may have to go back a step or two and make some adjustments. But do not leave holes in it or say this is good enough. Create a perfect process flow chart that allows anyone to do the job that has the required “skill set” and if they have questions they can refer to the operation plan for clarification. I look at both working hand and hand.
The scorecard is control chart that lets you input data to evaluate the process. It must be easy to do and part of the process. The data should not be labor-some to acquire or it simply will be put off in busy times. Data should be as instantaneous as possible, so that you can monitor a change downstream from the flow to see what may have caused the issues. Remember garbage in = garbage out. Hint: Your scorecard should check the consistency of the collection of data also.
The fourth step is what I call the savior step. Have a process in place or a go-to person that can react immediate to problems encountered, lack of data input, discipline, etc. This is your insurance that the process stays on track and is a success. It is not a onetime deal. If you decide one day that you cannot improve the process anymore, start working on the tools you use to create the process and refine them to dig deeper. It is a never ending quest. The result of this is that it is the only true competitive advantage that you have to stay ahead of your competition! So get there and stay there!
If you build a great control system the process will flow very well! When a process flows well you typically get great results! You can think of all the analogies to this statement but seldom can think of one that disputes it, can you?
If you are going to get to Improve, you have to know your math
Posted by: | CommentsIn the previous exercise, I just discussed the Measure phase of DMAIC and wanted to let you in on something, I left a little bit out. I think the average person hates math, especially marketing people. I only gave you a slight introduction, mostly for you to evaluate the process. Even if you only use “Tribal Knowledge” you will be better off going through the exercise. To get “Six Sigma” results you do have to measure and use statistical data, but to improve your processes, you can simply use Tribal knowledge, to an extent.
However, I will warn you, Tribal knowledge is not always correct. Observation of the numbers, without proper analysis, seldom works. You would be amazed at some of the scenarios I have seen where the obvious data does not work out to the obvious answer. But on with the show and how do you analyze something.
Typically, great discussion takes place on what should be measured. Make sure whatever you do has impact on the customer. You have a process map at your disposal; a set of measurements, so pick the one you understand best and there is an obvious problem with. This will make it simple rather than just trying to improve on the process.
You have two basic analysis to do, process and data. You can more than likely analyze either one first or even have separate parties do each. Remember the purpose of the analysis is to find the root cause of the problem. You will use these three steps, Exploring, Generate Hypothesis, Verify Cause. Support your decision utilizing one of the tools listed in the chart. There are more advanced steps, but this will give you a good foundation to start.
Some people make a few assumptions and go directly to the implement stage. One of my suggestions would be, if you did that, to at least experiment with your decision and analyze what change takes place. Testing is a big part of marketing and seldom used enough.
Data Analysis Tips:
-
Choose something that has different customer, segments, technologies, economics, etc. that are isolated. This way they will be easier to monitor the results.
-
Represents a significant or growing proportion of cost
-
Prioritize your objectives in price, speed, quality, service, etc.
Process Analysis Tips:
-
Processing time: the time actually worked on the job
-
Queue time: the time a job is waiting for the next operation
-
Wait time: the time a job is waiting on other parts of the process
-
Transport time: the time a job is in transit
Only processing time adds value, the other time elements should be reduced or removed.
Is this really possible to do in marketing? I think not only possible but mandatory in the coming years. And with the tools that are becoming available as we move further into the electronic age data will get easier to manage. Right now, we have more data than we know what to do with, but how are we analyzing website traffic or open rates, etc. Sophisticated analysis is becoming available to the masses. The understanding and implementation of this data is what will be imperative to have. Now you get ot go to the Improve step.
The start of this series of Posts was on How Much Value You Can Provide.
: DMAIC, Lean, Lean SixSigma Marketing, Market, Marketing, Six Sigma
