I live in East Sacramento, which is a small neighborhood just outside of downtown Sacramento. It is a quaint and established neighborhood, with beautiful tree lined streets, older homes, and lots of shops and restaurants within easy walking distance. Due to the easy walking distances, I do a lot of things right in my neighborhood. However, one thing I don’t do locally is my banking. I have an online bank which has physical bank locations only in Nevada. Thus, Automatic Tellers Machines (ATM) are critical to my banking needs. Beyond that, who really goes into a bank anymore anyway (cash isn’t even really that critical anymore, but I’m old school and like to have some on me at all times)?
There are two ATMs within walking distance of my home, each owned by a different bank. They are literally a block away from each other. However, the experience of each of these ATMs for the end user is miles apart. One of the ATMs has a customer focused design; I can absolutely tell whoever designed the functionality and steps of withdrawing your money out of the ATM was really focused on making this an easy and enjoyable experience for their customers. On the other hand is the ATM a block away, where the ATM experience is one of clumsiness and wasted steps.
Whenever possible, I go to the ATM with the customer focused design. The difference in the ease of use and experience is so great it makes me choose the better designed ATM. In addition, as you’ll see when we do some analysis, there is a step that is critical for me: the machine gives me my ATM card back right after I put it in the machine. This is critical for me because I’m completely absent minded, so if giving my ATM card back is the last step in the process, I will forget it practically every time. Well, not quite every time, but maybe 10% of the time, which is more than enough to make me tip my hat in the direction of the much better designed ATM. You may not think this is significant to the bank, however if a lot of customers feel like I do, then the bank is losing significant income on the ATM usage charges they are losing. In fact, I’m so anal about companies I work with having good processes, I will sometimes take my business elsewhere if I feel the alternate company will offer better processes. The reason is I feel this is an indication of a more effectively run company, and a sign of a management team who is on the ball. However, that is a separate blog to write…
So, being a process guy who is the leader of a process and decision company, I thought I might have some fun with this and apply some analysis to these two very different ATMs. My approach will be one of acting as if I’m analyzing the As-Is process to come up with a better To-Be process. I have the great advantage of already knowing the answer for the To-Be process, but there is still value in going through the exercise. During this analysis, I’ll be able to show some of the better features of IBM BlueworksLive (BWL). Now, it could be argued this is more about User Experience (UX) Design than Process Design, but I’m going to take the process approach since this serves my needs of highlighting some functionality within BWL.
In Figure 1, there is a BWL diagram for the “bad” ATM process.
Figure 1: As-Is Process
I’ve taken some liberties with the process. I have not included every possible gateway, path, or activity. However, this diagram is sufficient to show where the process has inefficiencies and was not well thought out, once we’ve done some analysis to highlight these areas. Plus, this is the path I typically take, and let’s face it, this is really all about me ;). During the As-Is process analysis, we determined which activities were value-add, and which ones had specific problems we would like to see addressed. In BWL we can use the Analyze feature to do both Value Add and Problem analysis. This lets me know what activities in the process are non-value add, as well as which ones have problems which were called out during the As-Is analysis.
Figure 2: Value Analysis
Figure 3: Problem Analysis
We will use the information from the Value and Problem analysis when we are designing the To-Be process. Let us assume we are now building the To-Be process. The approach we’ll want to take is to determine what problems we can solve, as well as possibly removing or improving the non-value adding activities. Let’s go through the Value analysis first. For the sake of expediency, we’ll only look at activities which are non-value add:
|Activity||Non Value Add Reason|
|Select Continue||Seems like the pin entry and a fast cash option could be on the same page eliminating a step/click. What value do I get from this?|
|Click Fast Cash From Checking||Why not just present the options for possible withdrawal amounts with another option to choose a different amount? Seems like a non-value added step as most of the time I choose $100 for my withdrawal, which is one of the options in the activity ‘Choose from Possible Amounts.’|
|Determine if Receipt Needed||This is probably very specific to me, but I never get a receipt. So, for me, this step is non-value add. Why not give the option to remember this setting?|
|Call Your Bank to Order New ATM Card||Not the ATM carriers fault, but certainly painful and non-value add. It could be argued this activity but doesn’t belong in the process, however I’m taking some license here.|
|Verify Card||Does not return card once card is verified as legitimate|
|Select Language||Doesn’t remember what language is associated with this card|
|Determine if Receipt Needed||Doesn’t remember whether user wants receipts or not|
|Return Card||Card should be returned earlier in process|
|Eat Card and Destroy||Requires user to go through inconveniences to get new card|
For the initial activities of Input ATM Card and Verify Card why not return the card right after it has been verified? I would imagine this is one of the ‘light bulbs’ which came on for the bank with the good ATM process. Once the card has been verified what value does it add to keep the card until the end of the process? In addition, due to me being one of the most absent minded people on earth, leaving the return card until the end of the process, in particular after I’ve retrieved my cash, opens the possibility of me forgetting my debit card and the ATM then having to ‘eat’ the card and destroy it. I would bet I do this at least 5% of the time I use this type of ATM. Granted, I am a ‘special’ person, but it is very painful when the solution is well within reach (no, the solution is not for me to get less absent-minded, that will never happen as much as I would like it to). This would look like Figure 4:
Also, in the inefficient process, note there is a Select Language step. Now, I’m not a genius, but I would be you there is a way for an ATM to correlate my choice of language with my card, and use that option in the future. I highly doubt I’ll decide one day I want my preferred language for my ATM to be Farcie rather than English. Thus, I’ll recommend allowing this setting to be remembered, and thus I have never have to choose that setting again. Of course, there could be edge cases where someone would need to select a different language, but that would be very rare and I'm sure can be handled in some manner.
Once the card has been returned, why should the user go through a bunch of steps to get to the point where they can choose the transaction they come to an ATM to do probably 90% of the time: retrieve cash? It seems a better solution might be to display a screen with the option to Enter Pin and Select Fast Cash Amount in one step. The user would also have the option to choose to do other activities at this point, however the overriding fact is the vast majority of the time the end user will want to withdraw one of the amounts of cash presented to them on this screen. So, we’ll make the change to the process and consolidate the activities Enter Pin, Select Continue, Verify Pin, Click Fast Cash from Checking, and ‘Choose From Possible Amounts’ into one activity. This greatly streamlines this process and removes redundant, or infrequently used, options.
Figure 5: Simplified sequence for first set of activities
Figure 6: Old process - sequence of activities
Our simplified first portion of the process is shown in Figure 5, while the old set of steps is shown in Figure 6. The activities in Figure 6 in red are those we have either removed, or made much simpler through consolidation.
As for the latter activities of concern in the As-Is process, interestingly enough the things we fixed in the earlier parts of the process have eliminated the need for all of the non-value add and problem activities. Previously, during the later part of the As-Is process, we had the activities Return Card, Eat Card and Destroy, and Call Your Bank to Order New ATM Card called out as either non-value add, required non-value, or a problem activity. Note the second half (denoted by the red rectangle) of the To-Be process in Figure 7; all of those activities are gone.
Figure 7: To-Be process wtih latter part of process highlighted in red.
Now that we have what seems like is a much more effective and efficient process, let’s go back through our value and problem analysis tables we documented earlier. This is a key step in determining whether the analysis work you have done has borne any fruit. If you haven’t solved the process problems, or resolved the non-value add activities, what is the new process doing for you? To quote Bill Gates “…automation applied to an inefficient operation will magnify the inefficiency.” Not every process will be automated, but in our case the process is an ATM process, so it is automated by default. Our previous analysis with a third column indicating whether it was solved or not is below.
|Activity||Non Value Add Reason||Resolution?|
|Select Continue||Seems like the pin entry and a fast cash option could be on the same page eliminating a step/click. What value do I get from this?||Yes. Consolidating steps eliminated need for Continue activity.|
|Click Fast Cash From Checking||Why not just present the options for possible withdrawal amounts with another option to choose a different amount? Seems like a non-value added step as most of the time I choose $100 for my withdrawal, which is one of the options in the activity ‘Choose from Possible Amounts.’||Yes. Consolidating steps eliminated need to have a separate activity to choose ‘Fast Cash from Checking.’|
|Determine if Receipt Needed||This is probably very specific to me, but I never get a receipt. So, for me, this step is non-value add. Why not give the option to remember this setting?||No. It was determined this was too specific to an individual process participant, and not global. Also, not a lot of gain by changing.|
|Call Your Bank to Order New ATM Card||Not the ATM carriers fault, but certainly painful and non-value add. It could be argued this activity but doesn’t belong in the process, however I’m taking some license here.||Yes. Returning card right after it is verified virtually guarantees card won’t be forgotten, thus bank won’t need to be called.|
|Verify Card||Does not return card once card is verified as legitimate||Yes. Changed to return card once verified.|
|Select Language||Doesn’t remember what language is associated with this card||Yes. Process now correlates card with initial language chosen.|
|Determine if Receipt Needed||Doesn’t remember whether user wants receipts or not||No. It was determined this was too specific to an individual process participant, and not global. Also, not a lot of gain by changing.|
|Return Card||Card should be returned earlier in process||Yes. Changed to return card once verified.|
|Eat Card and Destroy||Requires user to go through inconveniences to get new card||Yes. Returning card right after it is verified virtually guarantees card won’t be forgotten, thus bank won’t need to be called.|
Our new To-Be process is much more efficient and effective. We were able to solve most of the problem and non-value add activities called out in the As-Is process. We were able to reduce the number of activities from fifteen to ten, a 33% reduction. And, it just so happens this matches exactly to the process the from the ATM with the customer friendly process J. I don’t know if whoever built and programmed that ATM went through this type of analysis, but I would bet the process was somewhat similar. They did a hec of a job.
I just realized, as a side benefit of this little exercise, now when people ask me what I do, maybe I’ll have a story they can relate to as to how process improvement works. Usually, when I try to explain what I do, the questioner’s eyes glaze over. Hopefully this will help.