8 replies to this topic

[Scotty]

Hi. A quick query on customer complaints KPI calculation.

 

Currently I produce reports calculating cpmu/cphtu from complaints received against site sales to the retailer.

The retailer we supply has now asked for KPI's to be calculated through their 'till' sales and not our sales to them.

 

I have worked with various retailers in the past and this is the first time I have been asked to calculate in this format. Although I can see the logic behind this request, there is general feeling of being 'penalised' if the products do not sell/ wasted through store issues (for example - some of the store data I have seen seems to indicate a large proportion of waste due to products going out of date).

What are you views/ anyone else came across this or currently does this?

 

Regards

[Mike Green]

I have never come across a supplier being asked to do it -but it makes perfect sense for the retailer

 

But (for example) if it was bagged salad  in the UK -with a average of 40% being dumped by the retailer- or some equally perishable product- it's obviously gonna kill your stats!

 

it's probably time for a polite request to renegotiate the KPI's!

 

Mike

[Snookie]

 calculating cpmu/cphtu from complaints received against site sales to the retailer.

The retailer we supply has now asked for KPI's to be calculated through their 'till' sales and not our sales to them.

 

 

Am curious what cpmu/cphtu stands for. 

 

Am also curious why retailer would want complaints to be compared to till sales.  It could kill your stats as a vendor. 

[Mike Green]

Am curious what cpmu/cphtu stands for. 

 

Am also curious why retailer would want complaints to be compared to till sales.  It could kill your stats as a vendor. 

 usually complaints per million units/complaints per hundred thousand units

 

Guessing they want a more precise figure of 'complaints vs purchases' rather than 'complaints vs purchases plus waste' ?

 

If i was the retailer- I would want to do the same-as it gives a much more accurate picture of the occurance of issues with the product (the same issues may or may not be present in the scrapped product-so it potentially skews the figures)

 

(not sure I would expect my supplier to do it though!)

Mike

[Snookie]

 usually complaints per million units/complaints per hundred thousand units

 

Guessing they want a more precise figure of 'complaints vs purchases' rather than 'complaints vs purchases plus waste' ?

 

If i was the retailer- I would want to do the same-as it gives a much more accurate picture of the occurance of issues with the product (the same issues may or may not be present in the scrapped product-so it potentially skews the figures)

 

(not sure I would expect my supplier to do it though!)

Mike

 

Interesting.  I agree that would not expect supplier to do it. 

[bacon]

 usually complaints per million units/complaints per hundred thousand units

 

Guessing they want a more precise figure of 'complaints vs purchases' rather than 'complaints vs purchases plus waste' ?

 

If i was the retailer- I would want to do the same-as it gives a much more accurate picture of the occurance of issues with the product (the same issues may or may not be present in the scrapped product-so it potentially skews the figures)

 

(not sure I would expect my supplier to do it though!)

Mike

 

As an aside, I suspect these terms cpmu/cphtu (complaints per million units) are derived from Six Sigma statistical sampling measures of defect reduction or process performance: DPMO (defects per million opportunities).  

 

"six sigma" is a goal that one wants be be out six standard deviations between the process mean and the nearest specification limit (i.e. - airline industry flight landing "defects" are about 9 sigma out).

 

DPMO below is a statistical measure:

 

 

 

However, one could replace:

 

"Defects" = "recorded complaints"

"Number if Units" = "# of products produced"

"Number of Opportunities per Unit" = "# of things to complain about per individual product"

 

A "Complaint" Per Million Opportunity can then be translated into a sigma level.

 

So, with 1,000,000 products sold,

with 10 things to complain about per individual product, 

30 actual complaints

would be a "CPMO" of 3

(equaling 6 sigmas out from the process mean).

 

Just a thought.

-B

[moskito]

Hi,

we (bakery) are using the six sigma approach too, i.e. our overall target in complaints for each product is 3,4 complaints/1 million units (DPMO) -> 0,00034% failure/99,99966% failure free products.

 

Rgds

moskito

[Scotty]

Thanks for feedback.

 

Strange thing is the cuctomer already supplies us with a trading report with complaints mapped against till sales.

 

Regards

[Tony-C]

Hi. A quick query on customer complaints KPI calculation.

 

Currently I produce reports calculating cpmu/cphtu from complaints received against site sales to the retailer.

The retailer we supply has now asked for KPI's to be calculated through their 'till' sales and not our sales to them.

 

I have worked with various retailers in the past and this is the first time I have been asked to calculate in this format. Although I can see the logic behind this request, there is general feeling of being 'penalised' if the products do not sell/ wasted through store issues (for example - some of the store data I have seen seems to indicate a large proportion of waste due to products going out of date).

What are you views/ anyone else came across this or currently does this?

 

Regards

 

It sounds a reasonable way of calculating and a more accurate figure of cpmu, however I would continue to calculate as per your previous method so that you have some defence when asked to explain the inevitable increase in cpmu.

 

If they already calculate themselves it seems a bit overkill to ask you to do it as well but I guess that's another in a long list of hops to jump through

 

Regards,

 

Tony