iSixSigma

Defining CTQs based on Voice of Customer

Six Sigma – iSixSigma Forums Old Forums General Defining CTQs based on Voice of Customer

Viewing 12 posts - 1 through 12 (of 12 total)
  • Author
    Posts
  • #32761

    DMB
    Participant

    Hi,
    We made a Voice of Customer, the results are :
    – 40% would be satisfied to get an answer in 10 days or less
    – 60% would be satisfied to get an answer in 6 days or less
    – 80% would be satisfied to get an answer in 4 days or less
    – 90% would be satisfied to get an answer in 2 days or less
    -100% would be satisfied to get an answer immediately
    What would be the customers CTQ to define our sigma level ?
    – Immedialtely wouldn’t be relevant because we can’t deliver an answer immediately in profitable way. In addition our sigma would be zero and after a DMAIC project, it would remain zero after we reduced variation and median.
    – I am thinking about taking 80% as a starter (4 days).
    What are your thoughts ?
    Thanks
    DMB

    0
    #88038

    Bob Gouwens
    Participant

    We are a very customer inimate company, and our standard has been set at 3 days to return a response to a customer regarding quality issues.  Obviously some occurances require immediate response to determine if product is usable or may have to be reproduced, and other occurences require extended time to performe proper research and analysis.  We have learned from experience that if you provide the customer an immediate knee jerk reactive response without taking the neccessary time to research the issue thouroughly,  we have to go back to the customer later with another story (the correct story) which can create an integrity issue, and shake customer’s confidence.RGG RGG

    0
    #88041

    Sivakumar.V
    Member

    A suggestion would be to create 2-3(or more, depending on actual needs) levels of criticality and associate a response time to it because not all queries could be responded within a said time limit. Such classification would also provide some kind of guidenace on the actual resourcing needs needed so that those turnaround time frame could be met.

    0
    #88045

    Sinnicks
    Participant

    You did a VOC and are now asking about your project goal.  These are not the same.  The VOC determines your customers’ expectations.  You have probably also determined your current performance.  Your project goal is how much you think you can raise your performance in the context of this project.  It is not your standard forevermore.
    Pick a achievable, but stretch goal.  Pick a number, it is arbitratary.  There is no value add in it unless you do something to achieve the goal.  It does not affect your performance unless it is way too high, too low, or discussion about it keeps you from leaving the starting line.
    Mark

    0
    #88048

    Perryman
    Participant

    You’ve had some good advice so far.  Don’t jump to conclusions yet about what the customer wants until you have a better understanding.  You can use a Kano analysis to get a better idea of customer requirements.  This will tell you if you are wasting your time trying to improve performance of a criteria that will only provide marginal or no gains in customer satisfaction.
    You can check out the tools and templates bar on the left or drop me a note for more details.
     
    Cheers,
    Patch

    0
    #88053

    Ron
    Member

    Why did you go in two day increments?
    What you are saying is that it is acceptable to only satisfy 80% of your customers.. That would be a very low sigma value.
     
    Since immediate seems impossible ( Why?) I would run a continuous time clock ( 24 hours a day) record data which shows actually when a response is given. Often people think only in terms of days.
    It is apparent from your data that VOC says the sooner the better. So you target should be as fast as possible.
    You are struggling with at what point you generate customer dissatisfaction. What are the implications of dissatisfaction? Will you lose customers?
     

    0
    #88055

    Mike Tretola
    Participant

    There is more information in the data than you might currently be recognizing.  Since some customers might be completely satisfied with a slower response which I am supposing to be more cost effective you may want to do more profiling of your customer base and see what characteristics of the customer or the problem might warrant more immediate attention and which ones don’t.  You may be able to manage immediate responses for the 20% of customers that want it and make up for the added expense by slowing down you service levels for those that can wait longer.

    0
    #88068

    Klinton
    Participant

    Depending on the situation, the response time may or may not be a true value. The customer should always appreciate an immediate response. Although the solution may not be known at the time.
    – Some companies have focused on establishing the customer relationship. If the problem is so unique, and impossible to be resolved in a short time-frame, the customer will appreciate the constant feedback, and noticeable progress towards a solution. Both parties can work to improve the process, and agree upon the appropriate communication frequency.There have been successful companies in the past who have provided a “guaranteed” 24 hour solution. These companies worked to establish a “very” strong customer relationship; sometimes bringing the customer’s solutions inhouse, or onsite full/part-time. Working proactively, to prevent issues.Another company, with fewer resources, focused on the ability to reduce the number of calls. Defects were reduced, published solutions were provided to customers. By determining the root cause of the problems. These efforts would eventually a way to reduce the “time to resolution”. This provided a way for support team to focus on the more complex issues w/o context switching. This also provided a higher level of customer satisfaction.Cases failed where there was a poor customer/relationship. However, it is definitley possible to rebuild the customer/client relationship…Keep up the good work. I hope this helped…KL

    0
    #88080

    DMB
    Participant

    Hi ,
    Thanks for your great feedback.
    One of the comment was to pick an achievable but stretch goal and it could be arbitrary. I don’t really agree with this.
    If I set my goal at a sigma level (4 sigma for instance), it will be directly linked to the upper customer specification limit (UCSL) I will define.With the same process performance, I will have a lower sigma as a baseline with UCSL at 2 days rather than 4 days, or 6 days…
    If I choose let’s say 4 days as a UCSL and my process is 3 sigma. It means I have satisfied 80% (4 days) * 93% (3 sigma) = 74.4% of my customers.
    The issue is that by improving my process even to six sigma, I will satisfy a maximum of 80% of my customers which is pretty poor.
    The solution could be to set my UCSL to immediately (less than 10 seconds).In that case, my baseline would be 100% (<10 seconds) * 0% (O sigma) = 0% of my customers are satisfied.
    If my process entitlement is 1 day, after my DMAIC project, I would satisfy 100% (<10 seconds) * 0% (O sigma) = 0 % of my customers.
    This means my sigma level won’t show any improvement even if my process has improved.
    Choosing the correct UCSL is key. What I am looking for is the middle way.
    Here are some additionnal ideas:
    1) One way would be to set the UCSL close to the process entitlement level.2) Second way is to set my UCSL to immediately and forget about my DMAIC project and go for DFSS.3) I set a first UCSL for my project, reduce it for a second project…4) Fourth way (I like this one!) is to have a different UCSL for each customer.This means you ask at each customer what they find acceptable. If the time spent by your process for their specific transaction is greater than what they want, it’s a defect. This is a radical business change because you don’t use anymore averages but the real expectation for each customer. But you need a great reporting and measurement system too.
    DMB
    P.S. : Granularity is also a great point, thanks Ron, Go for the hour rather than the day !
     

    0
    #88085

    Al
    Participant

    Couple of thoughts approaching the issue of target Sigma level from a different perspective.  Initial question would be (without knowing your industry) whether speed of response is the true customer requirement.  Would suggest that a fast, but lousy response may not be as good as a slower but complete response.  Secondly, is there an opportunity to stratify the requirements – ie call type a) needs a 2 day response, call type b) needs a 4 day response etc.  Third consideration is baseline performance today and last question would be, is there an opportunity to benchmark with other parts of the company or the competition? 

    0
    #88135

    TS
    Member

    Hi bob
    We are also a very customer initiative company, and our standard has been set at 3 to 4 days to return a response to a customer regarding quality issues. As you mentioned we are also facing some delay for sending reply to customer due to various reason including delay in analysis. But now we have started monitoring the date and Review in every month for controlling the same in to 3 day.. What is the method your are following for monitoring above process. Whether you are converting this in to Sigma level for comparison. Can you help me for the same?.
     
    ts

    0
    #88163

    Bob Gouwens
    Participant

    TS
    We currently do not monitor and document time period for customer response, thus no sigma value applied. This is a good suggestion.
    We do however start the clock for response when the Account Mgr creates an Non-conformance report, and of course all response is dated, so we could process this data if needed. This process drives the Account Mgrs responsibility for reporting of the problem timely and the use of the quality system.
    Bob

    0
Viewing 12 posts - 1 through 12 (of 12 total)

The forum ‘General’ is closed to new topics and replies.