iSixSigma

Process sigma for FCR

Six Sigma – iSixSigma Forums Old Forums Europe Process sigma for FCR

This topic contains 13 replies, has 10 voices, and was last updated by  GP 14 years, 3 months ago.

Viewing 14 posts - 1 through 14 (of 14 total)
  • Author
    Posts
  • #23597

    kiruba
    Participant

    Hi there,
     
    I am working in a call centre industry wherein we have a metric called FCR – First contact Resolution. i.e., the number of issues that get resolved in the first time the customer contacts the technical support help desk.
     
    Note: If a customer calls us for the second time within 76 hours after making his first call then we don’t hit FCR for that particular call.
     
    High performing call centres following the best practices in the industry run at 85%. If we calculate the sigma value (Discrete data) for this metric it works out to 2.54 sigma (Macro Level).
     
    In a case like this though we have met the client expectation and are in par with the industry standards the process sigma value is less. But then, my benchmarking data or my client expectation is only 85%.
     
    In scenarios like this can we claim that we are at six sigma even though we are at actually running at 2.5 sigma as we already achieved out target.
     
    As per the basic definition of defect – “does not meet customer specifications or requirements” we have met the client’s target of 85%. (The service level agreement what we have with our clients)
     
    From a end user (External customer) perspective (15% of the customers whose issue are not resolved in the first time are dissatisfied so it is still a defect.
     
    Thanks in advance.
     
    Regards,kiruba.

    0
    #56311

    Ciaran May
    Participant

    Hi Kiruba,
    Dr. Deming has thought us to stop the practice of accepting poor quality so why does your industry accept that 85% FCR is OK. Look at the definition of Quality which is “meeting and exceeding customer expectations” The customer in your case surely expects a resolution when he/she calls I assume. I would. Therefore this means that 2.5 six sigma is about right and you have a long way to go to achieve six sigma for your process. Ask the question of who decided 85% FCR was OK! Did anybody consider the customer?
    Ciarán May
    e-mail: cjmay@indigo.ie

    0
    #56312

    Arne Buthmann
    Participant

    Kiruba:
    I would recommend not to change the Sigma level to 6 although you reach the SLA of 85% FCR. Reason is that in fact you still have 15% of the issues which are not resolved after the first call, i.e. 15% defects!
    For example in automotive industries customers require a cpk value of 1,3 (and higher) from their suppliers. That means, that they accept a certain defect rate but actually that doesn’t make the process more capable! Transferred to your case, your customers today accept a 2,5 Sigma level but that doesn’t make your process 6 sigma!  
    I think there’s no right or wrong. But to continue your Sigma calculation based on the CTQ “First contact resolution” makes more sense to me rather than to take the SLA of 85% as CTQ (Just consider that one day your customers may require 90% FCR!)
    Hope that helps
    Arne
     
     

    0
    #56313

    Hutauruk
    Participant

    Arne provides excellent advice and I very much support his position. You must work towards reducing defects further.

    0
    #56314

    kiruba
    Participant

    Hi Arne,
    Thanks for your valuable information.
    As of now from the business perspective ( i.e., in terms of making money) we don’t feel that it is a problem area which needs to be addressed immediately as we are already on par with induatry standards eventhough the defects are high (as the clients are satisfied)
    When we think of a going ahead with a project to improvise the customer CTQs eventhough we are running at a better sigma level in some of the other processes( like Csat), form business perspective I feel that working on that area will help us a lot more than this problem area.
    Would you agree with me on this?
    Regards,Kiruba.

    0
    #56315

    Arne Buthmann
    Participant

    Sure I do. One of the major golas of Six Sigma is to contribute to the company’s success. Therefore, projects should always start where there’s a ‘real’ and important business case! Arne

    0
    #56316

    kiruba
    Participant

    Thanks Arne.

    0
    #56317

    Dawn
    Participant

    I also come for the contact centre industry and really like the approach you have taken to reporting Sigma on FCR. About time someone applied this approach to the Contact Cetre Industry
    One of the things we have to consider here is why 85% is the target. FCR is based on an individual agents ability to respond to a question on the first contact. There are two factors here that make this a difficult process to control variation in. 1. There is high variation in the input – i.e. questions raised by customers can be on any subject, we can establish common patterns and train our agents, but we can’t control the variation in the questions presented. This part of the process is controlled by the customers.  2. We are talking about people’s ability to handle the questions i.e. the agents and so we have high variation in the quality of those individuals and their ability to answer, retain information, find answers.
    I think there is a lot more we can do by not accepting the industry norms – after all contact centres are a new industry with a lot to learn from others. However, how far we can push it is yet to be seen.

    0
    #56332

    Arora
    Participant

    Kiruba,
    While aspiring for higher FCRs than the benchmarked is good, the explanation that your process is only 2.5 sigma is not acceptable.Let’s not forget the the process capability index is a measure of           (USL-Mean)/3sigma or (Mean-LSL)/3sigma.This means that, the defect is the deviation from specification and not from 100% perfection(which in most cases is impossible).In simple terms, when you say 85% FCR is benchmarked, any thing less than that is a “defect”.Meeting 85% is six sigma(or more).
    It is good to aspire for higher FCRs.But, why is the benchmarking kept to 85%.You need to perform root cause analysis and hypothesis testing with the following reasons:Let’s take a computer mfr support division for example:
    1.In tech support, not all the problems are software related.usually 10%-20% of the issues are hardware related.which means, you will not have a solution handy but, the customer will have to place an order for replacement etc
    2.many software issues will also be involving shipment of CDs etc, which will again call for a longer time to close a case.
    3.certain cases cannot be closed on the first call.Let’s assume you have a computer and the tech runs “restore”CD on it which will take a couple of hours.Then the customer might have to call back again.
    4.Lets understand that many software issues don’t have permanant fix.While it works fine, when a customer downloads another incompatible software, it will corrupt the system and will need a call back.
    These are just few examples and you can actually do analysis on such reasons and build a nice Pareto to identify which of these are mostly taking up the rest of the 15%.But, in my opinion, to sustain 85% on a long run is itself a challenge.
    These are some of the questions you will ahve to answer before you are happy with 85% FCR:
    1.When you say you are meeting 85% FCR, what is the period you are talking about?.
    2.How do you measure FCR?(Is the souce your CRM tool or CSAT)?.If your answer is CSAT survey, then what is your response rate(no. of surveys/tickets closed)
    3.Is the response rate meeting the desired sample size sufficient enough to represent the whole population(all customers)?
    4.What is you correlation coficient between FCR and CSAT?(If it’s not positively correlated, FCR could have been obatined by data manipulation.which is quite common in call center industry)
    Hope his helps.Jes my thot.You can reach me @ krishnaa888@rediffmail.com if you would like to know anything else.
    Kris
     

    0
    #56351

    Ramya Krishnamurthy
    Participant

    Hi Kris,
    I hav not undergone any six sigma training but i’am working in a call center in Quality improvisation and working towards improving FCR. This a tech support process and i’am hoping to get some valuable suggestions to kick start with. This is a very important project for my career.
    Thanks
    Ramya Krishnamurthy
     
     

    0
    #56352

    Arora
    Participant

    Hi Ramya,
    Soory that was a pretty hetic wek.Coul’nt respond . It’s a good idea to improve CSAT since FCRs have an obvious positive correlation with CSAT.But, there is no “one size fits all” methods to address FCR issues to any organization.You will have to understand by root cause analysis what issues in your process/Program that causes lower FCR rates and address them.That is where your knowledge about statistical tools  help.When you go through DMAIC approach , you can definitely create a desired change.
    You can always email me @ krishnaa888@rediffmail.com.
    Best wishes,
    Kris

    0
    #56390

    Ramya
    Participant

    Thanks a lot Kris, it gives me confidence that i am heading in the right direction coz, we are working on CSAT. Our CSATs are based on feedbacks given by customers using IVR options using automated surveys and emails. We just hav a Q on the email format, which asks whether the call to the hlpdesk resolvd their issue. Based on the yes or no answer to this Q, we calculate the FCR for that agent. If u can suggest what data cud be derived from the reports and what action plans can be implemented to get a high fcr %, that will help me a lot.Thanks Ramya Krishnamurthy

    0
    #56655

    Psyc
    Participant

    One thing I have not seen in the responces is: What is the return on investment for inproving past the 85% mark. If it takes 6 months of training per employee to inprove it is it worth it. What is the cost to the business to have a low fcr?
    It is always a trade off. Is it cost effective to improve?
    Mike

    0
    #56656

    GP
    Participant

    Hi There,
    I am also from the Tech support Contact Center Industry and have come across the same problem in the past with FCR. What matters here is the FCR timeframe. As we all know it is not possible to resolve maybee 50% of the total of 20% hardware faults in one call or a two day time frame. So we changed the time frame to 7 calendar days and got more critical information on where the oppurtunity really was. We were also running at approx 90% FCR but when we changed the time frame we dropped to 50%. so we then did some RCA on this FCR failures and improved it with various actions and tools without any additional investment.
    After that when we took the 2-day FCR again just to see where we were as compared to the previous process we realised we were at 95% with FCR giving a R^2 value of 0.95 to CSAT. So maybee you could also work on the same lines to get a better hold over your process.
    Your client could be satisfied with the 85% at 2-day levels but are your customers really satisfied?

    0
Viewing 14 posts - 1 through 14 (of 14 total)

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