Customer Satisfaction

[Sinead C]

Hello all!

I am new to the quality scene and am just getting familiar with the standard and more importantly implementation.

I want to get away from sending customers surveys that they 1, cannot fill out more times than not 2, requires constant touch base and reach out points. I would like to compile the satisfaction data provided by customers through portals into some sort of normalizing formula that maintains our metric towards this KPI, eases the manpower necessary to obtain data, and enable a wider spread of input from customers. Is there anything wrong with having an internal function "normalizing" customer satisfaction metrics between customers for a single scored KPI?

 

[ScottK]

Hi Sinead - "Customer Satisfaction" is what your organization defines it as. If customer surveys don't work for you, don't use them. In fact, in most places where I've 'owned' the QMS, we did not use customer surveys at all. If you have an equation that is based on feedback through portals, direct feedback to customer service reps, complaint rates, even social media - that will work. Just be able to explain your method to an auditor when asked.

 

[Ed Panek]

I work in med device but if customer surveys arent providing the info you need I think there are other metrics you can capture.

1) On-time payment behavior. Some customers are always late. Im not talking about that. I am thinking trends on that. Work with your billing group to track that. Happy customers pay without needing 20 reminders.
2) They re up their contract with you without much hassle. They recommit for another 5 years with only minor haggling on price.
3) They are not sending an email about the cost its taking them to rework your products
4) Complaints generally
5) Complaints that require escalation internally - worse
6) Meeting behavior - Informal technical calls without contracts present. They are willing to talk to you about items other than cost.
7) They are willing to let you visit and access production to see how your product is being used.
8) Their audits of you. What have they said? Major NC? Is the tone warm or threatening?
9) Blame posturing. How do their engineers treat you when a problem pops up?
10) Sales feedback. Sales enjoys working with this account and is received in a friendly manner. Discussions are not tense.
11) When a problem comes up they call you first because they want you not to be blindsided. They dont draft a 20 page SCAR and email it to your entire QA group and sales group.

If you want you can weight each of these but these will be subjective but if you trend it that removes some subjectivity. Not all customers complain via literal complaints.

 

[Bev D]

‘Normalizing’ can be useful IF the data are real, measurable and objective. Please avoid numerical jabberwocky it will only fool you.
Remember what Deming said: some things that can be counted don’t matter and some things that matter can’t be counted.

There is nothing that requires a (fake) math formula. I had a Sr. Director of supply chain once ask me about a very complex formula for the supplier quality score. I tried to explain what this team was doing and how flawed it was. Then I changed track: I asked what were the 3 top suppliers that kept him up at night. His answer was swift and sure. I then asked him what suppliers were the 3 worst based on the ‘score’. (They were actually good suppliers and the 3 worst were listed as great.) ‘Nuff said. The score was eliminated.

 

[Sidney Vianna]

I would like to compile the satisfaction data provided by customers through portals into some sort of normalizing formula that maintains our metric towards this KPI, eases the manpower necessary to obtain data, and enable a wider spread of input from customers. Is there anything wrong with having an internal function "normalizing" customer satisfaction metrics between customers for a single scored KPI?

You posted this in the AS9100 sub forum, so, I will answer based on that. I believe that the IAQG is adamant about CB audits taking into account DATA and METRICS provided by the registrants customers as part of the planning and performance of their audits. I would pay a lot of attention to that aspect because, in the ASD sector, those quantifiable data, such as OTD performance and incoming product quality are paramount as part of the custsat equation.

The risk of having an “aggregated” index of customer satisfaction is the fact that it might mask some serious individual, granular issues with a few specific customers. It might give a false or misleading sense to management that our overall customer satisfaction performance is great while a few business relationships are on the brink of divorce.

ISO 10004 is a good guidance standard for the management of the customer satisfaction process. Consider implementing it.

Good luck.

 

[Kronos147]

I like the formula OTD + Qualty (DPPM) / 2.

Example:
OTD = 98%
Quality = 96%
98 + 96 = 194
194 / 2 = 97% = customer satisfaction

Surveys are still good to give the customer a chance to provide feedback, but is an ineffective method to determine customer satisfaction (IMHO)

 

[optomist1]

Yet another great timely thread, here is an excellent Quality Digest article by Bill Levinson re: cautionary considerations when using customer surveys. I found it enlightening...

Beware of These Common Types of Bias

Quality-related data collection is useful, but statistics can also deliver misleading and even dysfunctional results when incomplete.

www.qualitydigest.com

Hope this adds some value...
Optomist1

 

[Bev D]

I like the formula OTD + Qualty (DPPM) / 2.

Example:
OTD = 98%
Quality = 96%
98 + 96 = 194
194 / 2 = 97% = customer satisfaction

Surveys are still good to give the customer a chance to provide feedback, but is an ineffective method to determine customer satisfaction (IMHO)

You might like the formula but its numerical jabberwocky as the denominators are from different populations. It just substitutes ‘math’ for thinking. Math is a real thing. Its also not a measure of Customer Satisfaction. Sorry.
I do think surveys can eb useful to gain feedback but it is better to have real conversations with more than one group such as the QA or Purchasing group…
Customer Satisfaction is a complex thing and we shouldn’t try to over simplify it. Thinking is better than mathing.

 

[Ed Panek]

You might like the formula but its numerical jabberwocky as the denominators are from different populations. It just substitutes ‘math’ for thinking. Math is a real thing. Its also not a measure of Customer Satisfaction. Sorry.
I do think surveys can eb useful to gain feedback but it is better to have real conversations with more than one group such as the QA or Purchasing group…
Customer Satisfaction is a complex thing and we shouldn’t try to over simplify it. Thinking is better than mathing.

The moment anyone pushes on the "data" it gets very unstable. "Why did this customer rate you a 4 instead of a 5?" I really doubt they listed something measurable in any accompanying answer so the answer is "Its what they always give us"

 

[Sinead C]

Hi Sinead - "Customer Satisfaction" is what your organization defines it as. If customer surveys don't work for you, don't use them. In fact, in most places where I've 'owned' the QMS, we did not use customer surveys at all. If you have an equation that is based on feedback through portals, direct feedback to customer service reps, complaint rates, even social media - that will work. Just be able to explain your method to an auditor when asked.

Okay! Thank you for your insight!