iSixSigma

Data Gathering in Financial Services Account Opening

Six Sigma – iSixSigma Forums Old Forums General Data Gathering in Financial Services Account Opening

Viewing 10 posts - 1 through 10 (of 10 total)
  • Author
    Posts
  • #49983

    Terri Donahoe
    Member

    I am relatively new to the Six Sigma world and have started to work on a project regarding the length of time to open a new account at our credit union.  Does anyone have a suggestion as to the best way to collect data on how long it takes to open an account (short of sitting there with a stop watch); also be interested in how to determine our customers’ expectations as to how long it should take to open an account.  The concern is that asking them directly might not lead to a realistic response.  For example, if you ask me how long I would like to wait in a grocery store line, I will tell you 30 seconds, which, of course, is silly.  Finally, how many data samples should I collect in order to feel comfortable that the data is reliable?
     
     
     

    0
    #171618

    Stevo
    Member

    Good morning Terri,
     
    One of my favorite parts of being a BB is that I get to answer a question with questions.  For that I’m sorry.
     

    What was the catalyst that caused this to be a project?  VOC or management wanting to save a couple of bucks?
    Do you have a goal in mind? (Without baseline or benchmark data, if you have a goal, someone pulled it from their ___.)
    If you have no automated way to collect the data, yes, you will have to create a measurement system of you own.  However if you are inputting the application into a computer you most likely have a timestamp.
    Why wouldn’t you want to ask your customer what they expect?  It’s all in the questions you ask them.
     
    Collecting data is one of the most important tasks you should do, but first you have to understand what caused the platform to burn in the first place.
     
    Stevo

    0
    #171619

    Anonymous
    Participant

    Terri,
    What would be wrong with doing some observational research – during which you could measure the time?  I would hope that you would be more tactful than to pull your stop-watch out, hold it up in your hand for all to see and then start it when the customer sits down and stop it when the customer stands up to leave but you could sit in with a CS Associate for a few hours and take notes and gather some cycle time data…

    0
    #171624

    Juchniewicz
    Member

    Good morning!
    The catalyst was a core systems conversion which resulted in numerous customer complaints about the length of time to open an account.  Seemed to be far too many steps to do this – so VOC.  Management is reacting to that issue.  The goal is something to obtain from the VOC, but as I stated in my original e-mail, I’m not sure of how to phrase the question so I get realistic responses.  Any suggestions as to phrasing would be appreciated.
    I will check into the time stamp thing.  Thanks for the advice.
     
    Terri
     
     

    0
    #171625

    Juchniewicz
    Member

    Tact is my middle name, so of course pulling out a stop watch in plain view of everyone is not an option.  My concern was based more on having limited resources with which to manually time a process.
     
    If manual is our only option, any thoughts on how many timings would constitute validity?

    0
    #171626

    MrMHead
    Participant

    To expand on the stop watch approach – Do you have access to security camera tapes that display timestamps?
    Then, as long as you could identify the appropriate customers, you could FF through the tape to watch “Sit down” and “Stand Up” times  .. if that’s what you are looking for.

    0
    #171627

    Stevo
    Member

    Customer complaints are a great reason to start a project.  See if you can get a baseline count of how many complaints (back pocket).
     
    Customers are going to give you the best case scenario of they expectations, that’s a given.  Try to control that process by asking multiple questions such as:
     
    “Give me your ideal time you think it should take to complete out an application.”
     
    “What do you think is an acceptable time range to complete the application?”
     
    Both answers are valuable, sometimes you can’t delight the customer or even meet their expectations, but at least you make decisions “eyes wide open”.
     
    BTW – Since speed will most likely be you primary metric (could be # of complaints) make sure you add quality as your secondary metric.
     
    Stevo

    0
    #171628

    Outlier, MDSB
    Participant

    Terri,
    Stevo generally gives good advice on this forum, as he has done here.
    It seems you may have a kind of benchmark in place already though the data may be hard to get at this point. That is, use your old system for opening accounts as a benchmark. If complaints started after the conversion, then you might conclude that customers were at least OK with the time before. Is there any data on how long it took to open an account before the conversion? Is the old system still available in a technology lab somewhere to get some data?
    One simple way to compare old and new is to count “clicks.” How many clicks were there in the old method VS the new? A much more relevent measure of course is the time it takes now VS before. The good news is that since time is a continuous metric, you will need a relatively small number of data points (start with 20-30) to get a baseline of current performance. You could time 5 or six transactions per day over the course of a week and get a pretty good sampling. Make sure you take into account peak load times (lunch hour) and peak days (pay-day Fridays?) as you collect your data.
    As for gathering VOC, I agree, just talk to some customers and ask about their experience with the process. I wouldn’t rely on a customer’s direct time estimation, I would rely on some benchmark time, preferably coupled with high customer sat. data that shows customer are OK with the time it took. That may be hard to come by.
    In addition to gathering hard data on the time, you should also start looking at the new process itself. Are there any non-value-added steps in the process? Are there steps in the process that don’t need to be performed right when the customer is present? Are there steps where you have to re-enter information already collected somewhere else? (NVA).
    Good luck with your project. I work for a large financial services company too, so I feel your pain.
     

    0
    #171631

    Juchniewicz
    Member

    This is excellent advice.  Unfortunately, since the time to open an account was not an issue before our conversion, we never timed the process.  The data is no longer available.
    Your comments about the number of data points when collecting continuous data were especially helpful.  That gives me a good gauge.
    We are already thinking about the waste in the process, so your comments validated much of what we are planning to do.
    Appreciate your input!
     
     

    0
    #171632

    Juchniewicz
    Member

    This is exactly what I was looking for – help with the wording.  Much appreciated!

    0
Viewing 10 posts - 1 through 10 (of 10 total)

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