Need help with testing methods

Six Sigma – iSixSigma Forums Old Forums General Need help with testing methods

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
  • #44799


    I was wondering if anyone could help me? I currently manage a Welcome Process, i.e. once customers are acquired they receive several system generated letters, welcome packs and outbound telemarketing calls. I am looking to change the whole process, make it more efficient, cut down on comms and make them more compelling and most of all, trying to save costs by consolidating packs and comms.
     What sort of hypothesis testing, sampling etc etc should I use?

    Basically, I want to be able to prove that what I am doing is worthwhile and significant etc.



    test to be used depend by how do you measure customer feedback/satisfaction now. if you give more details, we can start to think about it togheter.



    We measure customer feedback through questionnaires within the welcome packs, one sent out at delivery of product and one +6 weeks of activation. We are also conducting some customer research within the Welcome Process. Main KPI is to reduce churn/attrition, maintain customer satisfaction so that they won’t leave us.
    My problem at the moment is that there are 4 welcome packs each customer receives and potentially another 2 on top of that. Along with this, there are several system generated letters that a customer will receive:

    Confirmation of products purchased
    Installation date of product (satellite Television)
    T’s & C’s
    Re-schduled Installation date
    Signature less contract letter
    1st bill explained letter
    All of the above letters are sent out for all 5 products, some customers sign up for multiple products. I want to consolidate this.
    The welcome packs received are as follows:

    Pre-installation pack – preparing the customer (5 days after being set up on system)
    Viewing card pack – transactional (1 week after system set up)
    Getting started pack – how to use the product (delivered at installation)
    Xsell and Upsell pack (+6 weeks of product activation which happens at installation)
    It is possible that they could receive the following packs too, dependent if they sign up those products:

    Broad band Welcome Pack
    Telephony Welcome Pack
    We also conduct Outbound Welcome Calls at +1 week of activation.
    My aim is to consolidate the letters into one Welcome Letter with all relevant information, the system has capability to do this, but it means a lot of back end analysis to create the different text cells for every scenario and the triggers. I would also like to consolidate the Welcome Packs so that a customer receives one a couple of days after being loaded on to the system, this will inform them of how to use their product/products and package/service. Then another pack at +6 weeks to cross sell products, packages and services that they don’t have.
    I hope that this helps.



    what you could do is : apply changes to your actual process as planned, for a defined sample of customer and check results with test. based on that, plan your hypothesis test.
    if you measure feedback through questionnaire, then you could have, at least, two KPIs : number or % of questionnaire received back (complete and/or incomplete) and a score value (I don’t know which, but it is not important).
    you know your actual KPIs values, so you can compare it  (before and after change) with of changed process (from sample),  running a test on frequency using z-test or chi^2 test (number of questionnaries returned)  and/or test on means (on score value)  z-test two sample for means or T-test paired two sample for means.
    hope to be of some help. if not, apologies, if so, plese let me know.



    As I understand from ur posting, you are basically working on reducing the processes of communicating the information to the customer. According to me it is a cycle time project.
    By consolidating the different processes, you are acheiving teo objectives:
    – Reducing cycle time of communication transfer to the customer,
    – Having more clarity in the information provided to the customer.
    You can collect data on cycle time before and after and do the hypothesis testing. By doing so, you will know whether your have been able to significantly reduce the time taken to communicate information to the customer.
    To check the effectivness of the information, you must have a survey done of customers before the improvement process and after improvement process. The sample of customer for before and after should be similar but not the same.
    You you should measure if you have been able to reduce your cycle time without significantly affecting the customer satisfaction.

Viewing 5 posts - 1 through 5 (of 5 total)

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