Hello all!

I will try to be as much specific as i can for the task i have been given. Please forgive any possible mistake.

Project : Complaining Management System in a reasturant chain business
Target : Improve System

First Actions : Gather info

Measure the complaints received from the phone center, DONE

Rank the complaints per store per year (since 2001), DONE

(the complaints are from customers who had their orders from DELIVERY or from TAKE OUT)

Write down all guest checks from 2001. The sum of guest checks is from Dine in Customers, Delivery Customers and Take out Customers. So i summed up only the guest checks per store per year concerning delivery and take out for the reason i wrote before. DONE

After that i was able to extract percentages. It is better trying to compare the level of complaints of two stores by that way. In 2001 for example the store with the highest number in complaints (77) it was actually in the fifth rank (sorting by percentage). And when i say percentages i mean "complaints / total guest checks (del+take out) X 100" DONE

Until here i think all are good exept the fact that we miss the complaints from the dine in (23% of the bussines)

Can you tell me please if i am walking the right way until now ? I know that this kind of job (food and especially restaurants) has many ups and downs but can anybody tell me if i can i use Trend Analysis using Minitab in this case to predict complaints? (i know that maybe it sounds funny but i will appreciate very much a straight clear answer. It is the first time i got involved so much and i have noone besides me to assist me.

Anyway i want to thank very much any kind soul who read this and moreover who will say a couple of words. :thanks:

Vasilis

Vasilis,
Is this thread related to your previous thread:
http://elsmar.com/Forums/showthread.php?t=8944
Some of us gave you feedback on the previous thread. Would you mind letting us know if the feedback was any useful to your project?
Regards,
Govind.

Hello all!

I will try to be as much specific as i can for the task i have been given. Please forgive any possible mistake.

Project : Complaining Management System in a reasturant chain business
Target : Improve System

First Actions : Gather info

Measure the complaints received from the phone center, DONE

Rank the complaints per store per year (since 2001), DONE

(the complaints are from customers who had their orders from DELIVERY or from TAKE OUT)

Write down all guest checks from 2001. The sum of guest checks is from Dine in Customers, Delivery Customers and Take out Customers. So i summed up only the guest checks per store per year concerning delivery and take out for the reason i wrote before. DONE

After that i was able to extract percentages. It is better trying to compare the level of complaints of two stores by that way. In 2001 for example the store with the highest number in complaints (77) it was actually in the fifth rank (sorting by percentage). And when i say percentages i mean "complaints / total guest checks (del+take out) X 100" DONE

Until here i think all are good exept the fact that we miss the complaints from the dine in (23% of the bussines)

Can you tell me please if i am walking the right way until now ? I know that this kind of job (food and especially restaurants) has many ups and downs but can anybody tell me if i can i use Trend Analysis using Minitab in this case to predict complaints? (i know that maybe it sounds funny but i will appreciate very much a straight clear answer. It is the first time i got involved so much and i have noone besides me to assist me.

Anyway i want to thank very much any kind soul who read this and moreover who will say a couple of words. :thanks:

Vasilis

Vasilis:

You are doing a nice job of analyzing this particular process. Remember that it is only one part of the answer you seek. You also need to figure out how to gather data for the dine-in experience.

Regarding where you are now with data analysis, beware of intense Minitab analysis too soon! Be sure to define your time parameters. You may want to compare year 2001 against year 2002, etc. You could make a run chart to see when the complaints happened. Is there a pattern? You may have to look at individual restaurants and complaints over time to see a pattern.

You know the percentages for each site for historical purposes. Now do you set a goal for improvement, and start gathering current data? When a complaint occurs, what is the process in determine root cause? How is it resolved? Can you create a pareto of reasons for past complaints and keep tracking it for the future? Then you can start tracking causes --

Good luck.

--Jodi

At first i have to say to Govind that YES the feedback that you gave me it IS a very nice way to make my work better. I maybe just run too fast (apparently wrong) to gather all these data and i didn't use the tools that you and the other people told me! I think that from now on i have to start using these tools (as Jodi tells me also and i thank her very much). Ok maybe i am eager to start using Minitab "Trend Analysis" i admit. I will start using simple tools to search for patterns.

Concerning sectors of problems the categorization has already be done. We have quality problems, Late problems and Wrong problems each of them focusing in different sectors.

The point (for the time being) for myself is to improve the quality problems (i am also a food techlologist and the orders were specific : "focus on quality problems" the other 2 sectors belong to others*). So i have to dig up from a LARGE pile of problems and see if i have any patterns.


*at the end of my project i will demand to get involved with these two other sectors cause noone handles them, and the worst is that they don't know how to handle them (ok i'm not genius but i have made the first step in all this thing)

Anyway thank you two, i will return with more data. I hope i'm not becoming disturbing. :thanx:

Try carving your annual data into smaller sub-groups, such as weeks, fortnights, or months. Plot a run chart ordering the data in the sequence that you carved it up into. Use a probability plot in Minitab to see if you have a normal distribution, if so calculate control limits, and continue to harvest your data in the same sub-group size.

Use standard rules for interpreting control charts, and hopefully you will have a useful tool and meaningful data.

Try carving your annual data into smaller sub-groups, such as weeks, fortnights, or months. Plot a run chart ordering the data in the sequence that you carved it up into. Use a probability plot in Minitab to see if you have a normal distribution, if so calculate control limits, and continue to harvest your data in the same sub-group size.

Use standard rules for interpreting control charts, and hopefully you will have a useful tool and meaningful data.


Yes, Martin, and Hi. Long time since seeing you here.

Also, I would like to add that a pareto chart might be in order here very soon. Martin's approach will tell you if you have a stable system. Instability could be caused by a number of reasons, such as a "rush" time. It may also be helpful to divide your data by specific problem type, as well (in a pareto chart). Do you appear to have one dish that gets a lot of "hits"? Or maybe complaints about slow service seem to be clustered around a specific time of day?

When you can find this type of information, it will point you to where your problem areas lie. Just remember, most of your problems will likely be because of a flaw in the system, and letting people be attacked with your data may make things worse. It sound basic, but it is important to attack problems, not people. One of the main things a food place has is atmosphere. The atmosphere in a restaurant with fearful employees could be as bad as a strange smell from the kitchen.

At first i have to say to Govind that YES the feedback that you gave me it IS a very nice way to make my work better. I maybe just run too fast (apparently wrong) to gather all these data and i didn't use the tools that you and the other people told me! I think that from now on i have to start using these tools (as Jodi tells me also and i thank her very much). Ok maybe i am eager to start using Minitab "Trend Analysis" i admit. I will start using simple tools to search for patterns.

Thanks. Nice to know that some of our suggestions were useful to you in your Quality Improvement Project. I asked the question because after some of us posted the replies, there was no feedback for such a long time. That thread kind of stopped abruptly without any closure!

Ok. Now looking at this thread, I understand that you have completed the collection of data of Dine in also. This is the main difference I see comparing the threads.

Look like you are in the "measure" stage already before "define".
I understand, your objective is
Reducing Customer Complaints by X %and
Improving bottom line by Y%.

While calculating Complaints/Total Guest cheques also try if you can calculate
Cost of Complaints/Total Guest Cheques.
Cost of Complaints could be:
Cost of the order replaced+Cost of processing the complaint+indirect costs if known.
If the Complaint cost is small, try Comparing the Complaints and Number of Customer visits charts to see if there is a decline in trend. Customer dissatisfaction spread fast..You may know the thumb rule.
One satisfied Customer tell 7 people,
One Dissatisfied customer tell 21 People.
If you had too many customer issues, this chart should give the hint that indirect cost i.e loss of customer is much severe than the small complaint cost.

I would have formed a team with experienced personnel from sections like Order receiving, Catering, delivery, waiter, shift manager, etc. These people can help you brainstorm and reduce the number Variable to analyze.

Example:
Currently you are analyzing the data at the Top level i.e Complaints/ guest cheques/store/year.
It could be drilled down to
Complaints/ take out/store/year and Complaints/dine-in/store/year
Further
Drilling per quarter
Drilling per region
Drilling per type of shift (day/evening)
Drilling per Weekdays-Weekends…
And many other possibility.

Unless you perform a team brainstorm, this list is difficult to bring down. Your team will be able to list the major causes that result in complaint and identify those that can be significant (by experience).

From that point forward, if you start your data collection and analysis, then you can understand the influence of significant variable that cause frequent customer complaints. Time sequence these variable and start to monitor closely.

Currently your metric at such a high level, by looking at the trend, you will not know if the upward , downward is due any specific reason and resolve. Too many variables and noise can cause this trend.

Let us hear from other professionals.
Regards,
Govind.
Anyway thank you two, i will return with more data. I hope i'm not becoming disturbing. :thanx:
Most effective way is to learn from your own experience.
Cost effective way is learn from others experience.

Thank you Govind for once more.

The dissapointing thing in my project is that there is no quality team and i am all alone. This is very bad i know but i can do nothing to change it (as you know the most difficult thing in improving prececess is to change the culture of the people...). Moreover for the time being i can't "touch" the other two major sector problems : Late (orders that were late) and Wrong (orders in which sth was forgoten). Having a quick glance in a line chart with these 3 categories, the quality problems were much more fewer than the other two.
However i do not give up but i am trying to be as good as i can to help the company i work for. So if i extract some meaningfull and usefull conclusions from this project i will be able to "demand" to get involve with the other problems as well.

You are right saying that i forgot the Define stage. For the time being i see a general trend (during 2001-2003) which says that dine in and delivery customers are getting fewer each year. On the other hand take out customers are getting bigger in numbers. But that's a general trend, a simple chart. The point is to dig up to see why's that.

I really would love to have a team cause the data i have is too much for 1 man to search. I have details which say the time of the order, the cost of the order, the rank of the customer (big, new etc), the cost of the compensation and more. The orders are around 7.000.000 and the complaints are around 5.000 so as you understand i will need a very big shovel :read:

Anyway all you here gave me very nice ideas which i am going to use very very soon. This time i will inform you to show you the follow up.

(what do you say about the number 0,07% non-conformances - concerning quality problems - as extracted from the numbers above? Do you consider it as a high or low number?)

See you soon, thanks again.

Vasilis

Thank you Govind for once more.

The dissapointing thing in my project is that there is no quality team and i am all alone. This is very bad i know but i can do nothing to change it (as you know the most difficult thing in improving prececess is to change the culture of the people...). Moreover for the time being i can't "touch" the other two major sector problems : Late (orders that were late) and Wrong (orders in which sth was forgoten). Having a quick glance in a line chart with these 3 categories, the quality problems were much more fewer than the other two.
However i do not give up but i am trying to be as good as i can to help the company i work for. So if i extract some meaningfull and usefull conclusions from this project i will be able to "demand" to get involve with the other problems as well.

You are right saying that i forgot the Define stage. For the time being i see a general trend (during 2001-2003) which says that dine in and delivery customers are getting fewer each year. On the other hand take out customers are getting bigger in numbers. But that's a general trend, a simple chart. The point is to dig up to see why's that.

I really would love to have a team cause the data i have is too much for 1 man to search. I have details which say the time of the order, the cost of the order, the rank of the customer (big, new etc), the cost of the compensation and more. The orders are around 7.000.000 and the complaints are around 5.000 so as you understand i will need a very big shovel :read:

Anyway all you here gave me very nice ideas which i am going to use very very soon. This time i will inform you to show you the follow up.

(what do you say about the number 0,07% non-conformances - concerning quality problems - as extracted from the numbers above? Do you consider it as a high or low number?)

See you soon, thanks again.

Vasilis The only way to eat an elephant is one bite at a time. Start with more manageable numbers.


Use the last year's complaints, since much may have changed since 2001.
Look at each restaurant as an individual process.
Do a pareto by category of each, starting with the one with the most complaints. Although it is 5th by % it is costing the most and generating the most disatisfaction.
You may find that there are a few problems identified causing the greatest amount of complaints (pareto principle). Work on these first to determine the root causes and corrective action.
In doing this you may find that some of the problems are common to many of the restaurants but the root causes may or may not be.

:2cents:

(what do you say about the number 0,07% non-conformances - concerning quality problems - as extracted from the numbers above? Do you consider it as a high or low number?)
Vasilis

Vasilist,
Please check out the other thread covering the same subject.
http://elsmar.com/Forums/showthread.php?t=8562

Iam not sure whether you meant 7% (0.07) or really 0.07%?
0.07 % looks very small percentage of non-conformance for any service industry.

Iam not able to comment either it is Good or Bad since I do not have any experience in Restaurant Business.

According to CQMgr Handbook,-US Office of Consumer affairs, Page 218: "For a product costing over $100, the Research result show that 9.5% of noncomplianants would repurchase”.

There could be many out there that were previously your customer, had a bad experience, did not complain, and did not return business.

While I would happy to see a small percentage of non-conformance in your business (restaurant), I would also be cautious to look at metric like Year after Year Revenue growth, and Customer Loyalty to validate that inference.

Regards,
Govind.

Vasilist,
Please check out the other thread covering the same subject.
http://elsmar.com/Forums/showthread.php?t=8562

Iam not sure whether you meant 7% (0.07) or really 0.07%?
0.07 % looks very small percentage of non-conformance for any service industry.


Hello Govind!

I am meant 0,07%!!! I know it's very small but the problems are still there!


Sorry for not responding faster. I am ''swimming'' through this "sea'' of problems and thank God your advices are my saviour. To tell ypou the truth now that i'm looking the problem from the start i changed my mind : it is bigger than i thought :eek: :(

But i don't give up. For a week now i use excels subtotals and the pareto Sooo much that my eyes are hurting (no joke). So i will be back , i hope shortly, to discuss with you and the other fellows my progress.

Btw thank you for the topic you suggested. I have in my mind to find a way to communicate with the writer.

Vasilis

Vasilist,
Thanks for the update. The reference I mentioned may be related to a US study. I was using it as an example to make a point. You may want to contact your country’s Chamber of Commerce to see if they issue such statistics on Customer loyalty, Satisfaction, by sector.

Iam already impressed to know that your complaints are only 0.07% !
Are you currently stable at 0.07%?

If you get an opportunity, please review:
http://qualitypress.asq.org/perl/catalog.cgi?item=H1090

Customer Satisfaction Measurement Simplified: A Step-by-Step Guide for ISO 9001:2000 Certification
Vavra, Terry G.
This book has a Software CD included. I find this book very impressive and relevant to your excercise.

Good luck with your efforts,
Govind.