Statisical methods for Customer Service Quality Assurance



Hello all,

I work in a customer service center and we are attempting to decide on the right sampling plan for our Quality Assurance program. I am not quite sure which one is appropriate or how to justify the method once chosen.

Basically, we have several different touchpoints (web based chat, email, phone) and low volume but many clients. We want to give a 95% confidence level, with 4% margin of error on both the client (external) and rep (internal) levels.

Any advice and/or directions to resources would be much appreciated.

Elsmar Forum Sponsor


Thank you for the statistical information - it helps out a lot. Well personally, I wanted to go with a straight 10% but my boss wanted to use sampling.

What other methods would you suggest we use?

Thanks again for the information


Fully vaccinated are you?
Rick! Thank you for helping out with this fellow's question and the others you have helped with as well. Your help is greatly appreciated!

Rick Goodson

The question is difficult to give a simple answer to. Statistically based samples are always more reliable than straight percentages. If you develop the Operational Characteristic (OC) curve for the sampling plan with a straight percentage you will see that as the population size decreases, the protection afforded by the sample gets worse. If you do not have a strong background in statistics I would recommend you look at a basic text on sampling or a basic quality control book with a good section on sampling before proceeding. My favorite is rather old but does have a current revision:

Quality Control by Dale Besterfield, ISBN 0-13-745232-2

Getting back to the original issue. If you are sampling as part of an ISO internal audit, you can not economically use statistical based samples unless you have an army of internal auditors. I am sure that Marc and some of the other contributors would agree. When I teach internal auditing I explain that there is not enough resource (auditors) to go around and therefore you have to take samples that are extremely small and not statistically significant. However, if you find a problem from a sample that is extremely small, it is usually an indicating of a problem worth investigating as the odds of your finding that problem were also extremely small.


Rick, thanks for your reply.

I have done some research into statistics and I admit that some of it goes way over my head.

I agree that as population decreases, the company will have to sample a LOT more to get the sort of accuracy we want.

Because our population is fairly low, I decided to use a longer time frame. As I stretch it out over a month and longer, the sessions to be reviewed decreases to a point where it is feasible to QC. Of course, we can't QC at the end of the time frame so we're using historical data to forecast and break up the QC on a daily basis.

That's been my solution so far, what do you think?

Thanks again for your input.


Fully vaccinated are you?
I teach my course the same way. I explain that there are so many specifics you have to set a sample but that they should understand that it would be overwelming to even begin to define what a statistically valid sample would be. As an example, for documents I tell them I suggest a typical sample size of 3 to 5 documents. No problems, go on your way. If you find 1 'reject', increase your sample size by 5. If you find another problem you are at a point where you will have to decide on the spot - can't be predicted as it will be a judgement call based on the specifics at hand. I emphasize that the sample size I recommend does NOT have a true statistical foundation.

Rick Goodson

Whether the use of statistical sampling is the right thing to do in this situation is a matter for debate. Never the less, the statistical side is relatively straight forward.

The formula for finding the sample size (assuming a continuous process that is in control) is:

n = sample size.
p = estimate of process percent defective.
Zcl= Z score of confidence level (remember
that this is a two sided estimate so
you have to use half the confidence
factor on each side, for 95% confidence
level you use 97.5% to find the Z
E = Allowable error of the process percent

Zcl x p x (1-p)
n = ---------------

In essence, the sample size is equal to the Z score squared times the percent defective time 1 minus the percent defective divided by the allowable error squared.


Process Quality Control by Ellis Ott, ISBN 0-07-047923-2, pages 76 - 79.

Modern Methods for Quality Control and improvement by Wadsworth, Stephens, and Godfrey, ISBN 0-471-87695-X, pages 173 - 176.

Hope this helps.

Rick Goodson

I hope you can make out the formula I just posted. Unfortunately the software makes it hard to accurately post a formula. Hopefully the text makes it clear.


Hi there ,
I need to understanding your objective of your question. What do you want to achieve with the sampling ? Are you looking for most preferred touchpoint ?

I am woking with several operating companies within my organization in their quality journey modelled after Baldrige Criteria. One of the area we focused on is "Customer Service" which deal with access mechanism (similar to your touchpoint). However, so far, we have not used any staistical method to determine the most preferred access mechanism but the following methods :-

1) Customer survey, though time consuming and subjected to realibility. This is to determine customer most preferred access mechanism (normally designed into the questionaire). Then do a Pareto chart
2) Data collection when customer call in and do a Pareto chart for most frequently used access mechanism
3) "do what others do" i.e. follow the market trend or competitor

However, the decision to provide which access mechanism depends on company's strategy and affortability. There is not standard way but to meet customers needs.

I could share more of my experience when I get a better understanding of your objective.

Hope this help

Thanks and Best Regards


[This message has been edited by lmfoong (edited 23 November 2000).]

Ray Culver

Good afternoon,
I have extensive experience in this area which perhaps I could share.
#1. For a customer service call center you must first start with a sample plan that is ALL trasactions, to get going on the project. You do this to determine the next stage which is understanding the "regular" sampling plan. This approach is Statistical Quality Control not Statistical Process Control. To sample your entire transaction base you need technology: The phones should be hooked to an ACD system, you web page should have a CMR engine driving it, your fax order transaction must be tagged as "fax" in your order entry modules of your legacy system, etc.
#2. I agree with Imfoong that you also must do customer surveying to determine their "touchpoints".
#3. Look at the wall where the company mission statement is hung. Half your touchpoints are probably implied in the statement.

In our company we handle about 650 tele calls a day, receive 1,000 faxes, 400 on-line orders.
We have all the technology as mentioned above to gather data on an hourly basis so we are at an advatage here...
We used customer surveying to find out what our customers wanted in service and we used QFD to structure the department(s), and determine our "touchpoints".
Here's some data to think about, because I think that your customers may be somewhat the same ours;
1. Customers want to make contact easy, fast, and user-friendly.
2. they want prompt, timely & responsive service.
3.They want their request to be handled once and only once.
4. Service must be error free.

We measure the call center on # in-coming/out-going calls, # adandoned calls, # Overflow calls, average time per call, # input errors, # abandoned webpage hits, etc, etc. This is done daily but also the data is used for strategic planning therefore is charted by month, quarter and annual.
We have found that we have violated customer requirement when:
a. There are more than 5 abandoned calls,
b. There are more than 1 Overflow calls (these are calls that are bounced out of the que to receptionist)
c. A call takes longer than 2min 50 seconds.
d. There is a mistake on 10 orders out of 1,600 per day.
e. We don't communicate back to them about ambiguous information from our webpage within 2 hours of receipt of request and within 3 hours of receipt of fax request.

What I'm suggesting is that in your original message you had a margin of error of 4%, which you may find is quite liberal! Did you mean .004%. We have found by listening to customers that the margin of error in most industries today is almost zero, due to technology, competition, and the demographics of society....

hope this helps.

P.s.Rick Goodson...great response for sample plan!!! This approach is quit necessary in a controlled customer service environment. But control in a dynamic department like C/S and Sales takes time, determination and's one of the hardest things to control and measure. Only now are we reading about pioneers in the area of quality and sales i.e Paul Seldon, Cas Welch.
Thread starter Similar threads Forum Replies Date
C. Tejeda Process validation of rework assembly methods (medical devices) Medical Device and FDA Regulations and Standards News 3
P Validation Methods of Machine learning and Artificial intelligence Pharmaceuticals (21 CFR Part 210, 21 CFR Part 211 and related Regulations) 10
J Adoption of automotive (AIAG) methods and documents AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 3
Sidney Vianna Informational ISO/NP TS 17012 - Guidelines for use of remote methods in conducting audits of management systems Other ISO and International Standards and European Regulations 0
Z Implement methods of Round Robin test Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
G Developing our custom test methods and test equipment US Medical Device Regulations 5
J Stock Traceability Methods, where to start? Manufacturing and Related Processes 0
A Recommended testing methods for weather stripping Manufacturing and Related Processes 6
H CAPA effectiveness verification methods for different types of CA Nonconformance and Corrective Action 10
R Statistical Methods for comparing test and reference product equivalence for quality attributes US Food and Drug Administration (FDA) 3
P Test Method Validation (TMV) for all Measurement Methods in Rec/Inspection Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
J CQI-9v4 Alternative TUS Testing Methods Various Other Specifications, Standards, and related Requirements 1
R Control Methods in Control Plan FMEA and Control Plans 5
M Methods used to label the package EU Medical Device Regulations 4
Crusader AS9100 7.5.2 NOTE Approval methods identified AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 8
A ISO 17025 - Methods validation and clients ISO 17025 related Discussions 3
Crimpshrine13 Laboratory Scope - Calibration vs. Test Methods - IATF 16949 IATF 16949 - Automotive Quality Systems Standard 3
Tagin Evaluating nonconformances for escalation using Bayesian methods? Statistical Analysis Tools, Techniques and SPC 2
BeaBea ISO 9001 Customer Feedback Methods - What has worked for your company? Service Industry Specific Topics 17
S High voltage testing - ISO 17025 - 7.2.2 Validation of methods and 7.3 Sampling ISO 17025 related Discussions 3
R IEC 60601-1 - Different methods of achievement of the isolation IEC 60601 - Medical Electrical Equipment Safety Standards Series 1
S IVF medical devices standards and test methods Medical Device and FDA Regulations and Standards News 4
S EU MDR Annex XIV - Clinical Evaluation Plan - What do these methods mean? EU Medical Device Regulations 16
O Software development plan : development methods IEC 62304 - Medical Device Software Life Cycle Processes 2
A Validation of Calibration Methods and Traceability - ISO 17025 ISO 17025 related Discussions 3
M Informational US – National Evaluation System for Health Technology Coordinating Center (NESTcc) Solicits Public Comments for Data Quality and Methods Frameworks Medical Device and FDA Regulations and Standards News 0
L SPC - Methods to collect data IATF 16949 - Automotive Quality Systems Standard 7
N Control plan evaluation methods - Which methods should be carried over from the PFMEA? FMEA and Control Plans 6
D Do Test Methods completed to a standard require validation? ISO 13485:2016 - Medical Device Quality Management Systems 7
qualprod Methods to Calculate Available Production Rate ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 0
M Methods of reducing/eliminating the systematic component of error General Measurement Device and Calibration Topics 0
S Methods to reduce relative humidity in Datacenter IEC 27001 - Information Security Management Systems (ISMS) 3
K Alternate Control Methods (IATF 16949 Cl. IATF 16949 - Automotive Quality Systems Standard 1
H ISO13485:2016 Measurement System - Methods General Measurement Device and Calibration Topics 4
armani Non-destructive testing methods are special processes? Inspection, Prints (Drawings), Testing, Sampling and Related Topics 18
S CAPA Investigation Methods Nonconformance and Corrective Action 3
armani Monitoring and Measurement Methods ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 11
R OSHA - How often an employee needs to be Re-Trained on Safety Methods Occupational Health & Safety Management Standards 4
Ron Rompen Sintered Metal Machined Parts Rust Removal Methods Manufacturing and Related Processes 8
B Gage R&R MSA - Variation, Tolerance, Variance by different methods IATF 16949 - Automotive Quality Systems Standard 23
X Acceptable methods to store and archive records ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 7
S Rapid screening methods for antibiotic residues in Chicken Meat Food Safety - ISO 22000, HACCP (21 CFR 120) 1
J Software and Methods for Tracking CAPA (Corrective and Preventive Action) items US Food and Drug Administration (FDA) 3
A What are methods and tools to measure Workload analysis ? Human Factors and Ergonomics in Engineering 1
Q Management Review - Alternative methods Management Review Meetings and related Processes 20
Gman2 Control of Records - Lets talk about modern methods Records and Data - Quality, Legal and Other Evidence 5
R Gage Identification Methods in a Job Shop Calibration Frequency (Interval) 8
K Control Plan vs PFMEA Control Methods FMEA and Control Plans 1
S Consolidation of Testing Methods Using Minitab Software 3
G Visual Inspection of Machined Parts - Best Methods Inspection, Prints (Drawings), Testing, Sampling and Related Topics 8

Similar threads

Top Bottom