Toward the Electronic Office (1981)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

OFFICE SYSTEMS INNOVATIONS IN THE BANK Louis Mertes Continential Illinois National Bank Over the past two and-a-half years the Continental Illinois National Bank has converted about 25 percent of its employees to using electronic mail and information retrieval. What I will cover today is what our view is on some of the problems and issues that we have encountered, how we see our progress, what we are trying to accomplish, and also some of the specific things that we have implemented. First, to give you a general overview, I will remark on what has been happening in the world of automation over the last few years. During the '60s our view was that almost all automation efforts were pointed towards the production areas. Nothing was oriented toward the computer people in terms of making their job more efficient or administrative and professional areas. If you do not believe that, see how effective they were in installing systems during those years. The following ten years we again continued to move toward the production areas, did a little more in the development areas in terms of up-grading the quality of programming languages and trying to improve our own operating environment. We also did some things in word processing, dictation and a few managerial type of reports. Our view of the future, the 8O's, is that professional productivity will be our main area of involvement. Activity is going to be in the administrative and professional areas and the real productivity gains are going to be there. The reason we have gone ahead in this automation area, properly termed as office automation, is to improve professional productivity and keep the focus on that; not on making the office more efficient in terms of word processing or getting the secretary more effective, even though as a by-product those things do happen as well. About two and one-half years ago when we started implementing some of the electronic mail and information retrieval systems, three members of my staff and I analyzed how we would feel operating in an environment that we were going to suggest and impose, or help guide, 7l

our entire organization to move toward. At that time, I threw the desk out of my office, after deciding that handwriting and similar archaic things should no longer be done. Within the first 90 days, I concluded I had lost my mind and was wondering how I was going to back off. What occurred over those 90 days made me and several of my managers feel the magnitude of the change we were going to impose on ourselves and that we were going to suggest as the organizational response to office automation. In fact, I think one of the more important things about this area of automation is this is the first time, at least in our organization, where the decision-makers are making decisions to automate themselves. If you look back, when someone was automating a payroll, the people who were making the decisions were saying, let us automate the clerical people and then they complained they had no choice but to continue on because once the payroll system was converted they could not go back to the manual perspective. However, now we are asking people to make a decision to change entirely the way they behave and also provide them the option of not using the system after they started using it. So, we had a whole different arena, at least from our perspective, in terms of behavioral ramifications and also on retrenching staffing use once the conversion has taken effect. Our managerial belief right now is that professionals can save time if equipped with effective tools. I believe I personally save at least two hours a day in doing the same kinds of activities I used to do manually. For example, I have a dial-up terminal set into a coffee table. It pops up electronically, so it goes away and you can not see it when I don't need it, I have a picture phone on the desk to do teleconferencing with my staff. I do dictation through a telephone to a remote dictation center. The center then transmits it through a word processing unit to our CPUs. I retrieve it and look at a draft copy before I release it on the terminal at my desk or at home, or anywhere I can get to a terminal. I have an answering device on my phone to pick up my messages so that I can deal with verbal mail and also pick up messages from remote locations. The whole orientation in our organization in moving towards office automation is location independence. How can I operate in a fully functional fashion being anywhere I choose to be? This is a very interesting question. It opens up a whole new opportunity of suggesting where should you be to do your job. Today, we are so tied to our physical location because that is where our phone messages stack up; that is where our mail is mailed to. We haven't really thought through where we could more effectively operate if we could have full support wherever we are. 72

So, it really opens up the question of, what is an office and why an office? It directs our focus on location independence and also on testing and trying pilots that find out how people feel about not coming down to work a couple of days a week, or doing word processing in the home by dictating to a remote word processing terminal in someone else's home. So, what we said was, before we go ahead and charge on with all these crazy ideas, what exactly are we trying to accomplish? We geared them down to four potential areas of activity in our office which are obvious once you address them. First of all, all of us do reading and writing. We said we were going to address the reading and writing function of how you create the text, how do you process it, how do you retrieve it when you want to look at it again, where do you file it, and also how you get it approved. So, we were going to address the reading and writing function, but stay away from anything specific that people do in their reading and writing process. All we were going to do was address the function. The next area we were going to address was talking and listening. The issue here is that in the talking and listening arena we get involved in something called a telephone tag. The issue is I will call you and then you will call me back and then I will call you and you will call me back. We call this process telephone tag in our organization. A lot of studies show that it takes three or four phone calls to make one complete phone call. The focus was, how can we eliminate that nonsense or that stack of messages all of you are going to get when you go back to your desk today or tomorrow. The other one was traveling to and from meetings, even in large buildings. Everyone in our organization used to believe that you had to be going to Europe or New York before you wanted to consider teleconferencing. However, I have a group that is about three miles away from me and the travel time is an hour round-trip, portal to portal. The issue was, how can you provide some teleconferencing capabilities or some kind of ways to eliminate those one-hour trips during the day, which is wasted time. We may believe we are studying as we are walking down the street, or thinking, but it really ends up not being very productive activity. The fourth area concerns something that we know government does not do. What we tried to do was reorient what was office automation. What we were really saying was let us address the work patterns of the professionals, people engaged in reading and writing, talking and listening, et cetera. It is not about gadgets, even though a lot of things we tend to talk about concern what kind of word processing gear do we want, what kind of manager's work stations? 73

Our view of a manager's work station is cheap is good. If you can build something for $400 or $500, that's great. Every time I talk to a supplier, they say, are you willing to spend $l0,000 or $l5,000. I say, you are absurd. Think $500 to $l,000 and maybe we can get some progress made. Our objective is to get a device that can handle the communication of all the information on everyone's desk. That means you don't want a financial study every time someone is hired on whether they merit the opportunity of having a $20,000 to $30,000 work station. What you want is something inexpensive, like the phone, to be put in place. So, we next addressed two things on how. First, how technically were we going to deal with the situation. Many of you may deal with suppliers on a regular basis. They are a very interesting group to deal with. Some people think that anyway. But, what we were looking at in l977 and l978 was some guidance on what suppliers were talking about in terms of office automation and how to pull together all the technology. They had their own idea and most of them communicated only those directions that they themselves were supporting. As a result, what we said was we need to get our own technical philosophy straight so the next neatest vendor idea that hits the street does not redirect our direction or impede our progress towards an end. We needed to really focus on the function we are trying to provide to the professional and put the technology and the computer in the background. So, what we said were two things. We want to focus technologically towards a central library, everything stored in one logical room, where you have categorized all the things that could be accessed from one location not much different than a library in your home. Therefore, if you wanted a math book you look in the math index. If you wanted a physiology book, you go to the physiology index, but you do not have to worry about where you have to go. It is all contained in one central library. Second was a notion of a universal terminal. I think someone mentioned that just earlier. One terminal with a dial-up capability, that could access any information in the central library at any time through any telephone. So, those were the two technical concepts we incorporated in our overall approach. We have allowed any supplier, who can help us in this direction, support our activities since then. The next question we addressed was how to package the capability. As the gentleman from Avon highlighted, we all have backlogged computer projects. If I brought this overall project two years ago to my management staff and said, I would like to spend $5 or $l0 million to study office automation and make the offices more efficient, they would say very clearly, go back and do the work you are behind on, quit fooling around with the future stuff. No one knows if there are any true productivity gains inherent in this process anyway. 74

So, what we did — and we concluded that that was the kind of response we would receive — we used the old military game of scrounge; cheap is good, don't assign a lot of people because that just means you up the cost of the process. We assigned two people initially to the process. We said, find out what you can find out there. Subdivide it into mini-projects not an overall whole, as long as you are aware of how you want the puzzle to be put together. That is the exact approach taken. We moved ahead on information retrieval, audio mail, electronic mail, graphics, and analysis. But, we did them separately, without saying, here is our overall strategy on how to improve professional productivity. As we moved ahead with quick pilots, we then got in the process of having people' justify them from our user perspective in terms of putting the pieces in place. Other things we did just expanded in the organization as long as there was a request to get involved even though they could not justify some parts. The last step was, how to implement and where? We approached it not too much differently than most marketing firms approach introducing new products. Get to the early adapters, people who would be receptive to change and would profit from it. Do not try the new technologies or ask people to change totally who reject change in the first place. We recognized we were going to have enough complaints from the people that wanted to try it, that you had to stay away from those who had no interest in the first place. Not only did we want the early adapters, we wanted those early adapters that we would view as supportive of what we were trying to accomplish, because it is so easy to toss the whole thing down the drain and find it not worthwhile to do. We also subdivided different pieces of the pie among the early adapters. So, if someone went sour and did not like one piece of the action, we could move to another area and test it elsewhere, and really just separated the whole project into a lot of small pieces and moved ahead. The first item that we started to install was audio mail. This is a very simple thing to get on with. It was one of those things where three of us were sitting and chatting about the problem of telephone tag and said, how do you get at this? What we did was install answering devices which each and every one of you has run into sometime in your life and hated. After the beep, please leave your name and number. We said, let us take this thing that has been around for l5 years and change the way we look at it. What we said was, view it as your option to talk to the secretary or a receptionist who might take your name or number, or let the phone ring for ten times. The last option was you have an answering device on which you could dictate a full message. Now, the 75

devices we installed hold up to 30 minutes. This even allowed for some of our more verbal people. We also had fast forward so we could eliminate listening to overly long messages. Look at that process and, rather than being irritated by having to talk to a machine, view it as an opportunity to dictate a message because you could not talk directly. Quickly we put in l2, then we put in 38. Within my own management structure we went right from myself and covered the first l2 guys reporting to me and then went down 38 to include 38 people. We found the following things. One is, 65 percent of the messages were being picked up on the answering device. It scared us a little bit initially but we checked with some consulting firms and found that a manager or a professional is away from his desk or on the phone something like 25 or 30 percent of the time anyway. Sixty-five percent going into the machine was still maintaining the behavior that we had expected. That was, they continued to answer the phone no differently than they would have before, not get people sensitized that they are not going to answer the telephone anymore. Secondly, we put it on the second ring so that no secretary or receptionist intervened, ever. A caller could count on leaving a message. Of the messages called in on the device: in excess of 50 percent were either one-way or two-way complete messages. What I mean by one-way complete messages is, you called me and left a full message and there was no need for me to return the call — it was just information, one-directional — and there was no need to call back and I knew exactly what you wanted to communicate. For example, my data center manager, after a bad evening, preferred talking to the machine than to me. The remaining complete messages were two-way messages where you would call me and leave all of the questions you had on your mind. Again, think you are dictating a verbal memo. I would call back and you also were not at your desk — which is a very high likelihood — and I was able to answer all of the questions. What we found was that over 50 percent of the messages that were picked up by the machine were handled that way, and that that percentage grew as people got more comfortable in the process. The thing we did find, though, was that people went through a learning period and people that installed the answering devices had to almost be crusaders in a certain way. The first time an individual called and ran into the machine — he would say, how can anyone in a bank have an answering device on their machine and hang up. We also had some choice words left on the machine because of previous experiences. 76

What was more important was that if you walked them through and said, last time you left only your name and number, however, here is what we are trying to accomplish; the next time they would really try. What they would do is leave their name, number, and usually, the subject, but never give you a full dictated memo. You would have to be the crusader for the next step and say, here is what you forgot to say. It works. It really does. I have no more friends, but it does work. So, what we have done is expanded this. It really works well and it eliminates the stack of messages that you inherit by not being at your desk. It allows location independence, because now from here or from wherever I am, I can pick up my messages. I have now left town and behave as if I was still in the bank. I pick up my messages, people view me as still being at the office because there was no face-to-face contact required. It does work. I think some of the suppliers are coming up with some interesting things to further enhance the shortcomings of just an answering device. They are simple to install and you can move quickly without a major study. The devices cost somewhere in the range of $350 to $400, so they are not a major expenditure and progress can be made quite well. Another important thing was grouping and installation. It turned out that if you spot the devices around an organization they do not work very well. What you need to do is go down the managerial chain because it represents the calling patterns you are concerned about. If people calling you call infrequently enough you do not really learn the process or walk them through the three-phase approach of getting to the point where they leave full messages. Another point is you learn to prepare to give a full message when you make the phone call, especially if you know they have an answering device. Now, when the answering device is turned off or the person answers the phone you really feel frustrated at times because you are not able to execute the full communication without interruption. Other areas that we worked on were dictation and word processing. We think that moving people into both of these directions is critical to getting everything electronically captured. We, at the bank, have had dictation going fairly aggressively and in general we have been moving towards remote dictation using the telephone to dictate to a remote word processing center. What we have done in the last two years is hook up several of the word processing centers within the organization with communicating devices to our data center so individuals can electronically move the data to an end-point terminal rather than being required to pick up some paper at a specific locations. 77

We have had two pilots that we have been working on in addition to many word processing centers located around the bank. One is a pilot with two people in their home. We feel there is an excellent potential of using qualified people that are tied to their homes or choose to be there. We have made the underlying assumption that equipment will be cheap enough that you become indifferent to whether it is being used eight hours a day or six hours a day. The important thing is the quality of people that you can hire to perform the task not where they are located. So, we have installed several devices in people's homes. I can pick up a phone, dictate to a word processing remote dictation center. The person in the home can pick it up, key the information, transmit it to our CPU and move the information. So, what you really have is you and even your support location are independent. I can dictate from here in Washington to someone in a suburb of Chicago and she in turn can retransmit, or make available that information in my mail system where I, again, can read it, edit it and resend it and ask for further development of the process. You end up being totally independent of where she or the other people are located in performing their tasks. The other pilot is where we said, if the central center equipment is going to be expensive, you want to put it in a satellite away from your main source area. We located a word processing center in a suburb of Chicago. What you do is dictate to that remote area. We work the machines multiple shifts a day. Both pilots are going along well. We are finding some things that need to be improved in terms of the hardware to make it a little more smooth. But, the process is working very well and it is really making some behavioral changes with the groups around the organization. The other area that we tried was electronic mail. When we installed our terminals, I ended up frustrated because I could not key at all. If you read the Wall Street Journal, we also have a manager in our bank that thinks keying is bad. The conclusion when I eliminated my desk was, how am I going to now handle these things? I handled them in two ways. One is I dictate remote, but on the other hand, I also bought myself a teach-yourself-how-to-type book. It takes about ten person-hours to get yourself to type faster than you can write. I don't mean with all the capitals and indentations and punctuation, but to replace handwriting, because that is the objective. Another interesting by-product is I am now able to read the things I created yesterday. So, when we went to electronic mail we did the same thing as with audio mail. We took another group, installed it down the management chain, and said let us start putting up documents electronically. It is starting to replace telex in our organization. 78

Our European heads love it because as they are moving from country-to-country, traveling, people can reach them and send them information no matter where they are located because any time they stop at one of our European branches they can get on a terminal and pick up their mail as if they were back at their head office. One of the problems we are running into in the European relationship — and some of you may be aware of it — is the transborder data flow issues that are happening between the countries. People are raising questions on can you really transmit information in and out of different countries. Restriction of transborder data flow, from our perspective, would be a disaster and hopefully people here or someone in government is taking a very aggressive stand to help us in this area. The mail system can be entered in three ways: remote dictation or key it yourself; or dictate it or handwrite it and hand it to a secretary or some other clerical person to enter the information. It moves electronically so distribution is instantaneous. We file it or separate it. It is not very different than what you do with a paper-based system. We sort it into different categories no different than your in-box may be sorted today. We are now doing approvals on the mail system so if I want something approved through four levels of the chain of command someone would enter the four people that it has to be routed to. It is automatically routed to the first person on the list. The second person gets it as soon as the first person approves it. We are testing how people will operate within the work environment without seeing a physical signature. It seems to be working within a small group right now. We expect to expand it to several thousand by this year-end with no trouble at all. Traffic is separated by primary recipients; in other words things directed directly to you, and things you are copied on. We have bulletins so I can broadcast to the 800 people that work for me in the organization with just one memo and one command, "my staff;" and it goes out to the 800 people instantaneously. So, all of a sudden I can almost beat the rumor mill. We just break it down into a lot of categories. You ask for display and see what you have got in your in-box. The other thing is that you can browse or read memos. You can comment. In other words, like a paper buck slip, you can just comment by typing in a sentence or two and re-route it, routing to multiple people. We allow for the use of nicknames, so you can create your own nicknames of people and don't have to spell out full last names. I understand some of the nicknames used are quite interesting in terms of other people. You can staple the document electronically right away, so that when you comment you can just staple it and that document goes with it, no different than what you would do in a copying 79

environment. Using "CBS" command, you can comment, send it back to the person that sent it to you and staple a document. So, just with the simple three letters, you can readdress the memo to the person that sent it to you and it is sent back instantaneously. The "send" is straightforward. We have the approval process, which I mentioned. We also have "registered" mail. I can send anything to any group of people and it will tell me when they have looked at it. All it does is keep a master listing of all the people I have sent it to if I send it "registered." It is no different than registered mail. When you have looked at it, it will give the time and date so that you can keep track of what is happening on important documents or where they are. It turns out to be an important process to the approval process, because as you are routing things through the chain of command or around an organization you always know where it is, who has read it and approved it last, and whose desk "it is sitting on," without having to run around or spend all kinds of clerical and staff time trying to find out where the document is that you are trying to get signed off. We are also expanding this notion on mail. One of the important things we have found in the mail system is you have to go down multiple levels of the chain of command. Initially, we started with the head of our international banking department and took only the people reporting to him and did not go down two or three levels. We found out if you do not go approximately three levels down the chain of command and sometimes four you do not have the worker. When you only go down one level, no one in that level is creating anything, so nothing is being transmitted. So, the important thing in the installation of mail is to go deep enough in the organization so that those who create documents are doing it electronically. Then you can keep mail electronically rather than having people read documents that are on paper. We are also expanding this. We are linking word processing to our computer systems and we are coming up with computer-generated mail. The question is, why should all the mail be created by other people? Now, rather than creating exception computer reports that are specifically formatted, let the mail system be your in-box and when the computer is programmed with limits or guidelines on report suspenses let it cut a memo and send it to people saying, "Hey, you have an expense statement that is overdue, you have this overdue." We are looking into starting to expand that notion. We expect that the area of computer generated mail will probably be a larger function over time because there is a lot of exception and control reporting that is desired. Another point on the mail system we debated was what ought to be in a mail system. We concluded that we could debate forever on how each of us handles our in-box and really, who cares. Finally, we said, let us get something done. A young lady raised her hand and said, "I will get up a pilot mail system in six weeks." I said, 80

sold. And, she did it. Six weeks later we started with about 60 people. That system now has been scrapped. We only got up to l200 people before we scrapped it. But, we scrapped it and enhanced the follow on system since then. But, we did get up to l200 people using an electronic mail system that was thrown up very quickly. We found that it is more important to get people using electronic mail that are willing to use it and that give you meaningful feedback on how it should be improved, rather than theoreticians sitting back telling you how you should do your mail and nothing happening because you are not learning anything directly. IRIS: another area that we investigated was information retrieval and the big desire to have information at our fingertips. The underlying product in IRIS is IBM Stairs-Virtual Storage (VS) which is a text retrieval package. We contacted IBM, some people flew in from New York and we chatted with them. They said, by the way, we have this product that is clearly an overkill. You won't want it. Right away we started paying attention. When IBM says I won't want something I figure I want it. They said it is a heavy text retrieval system. If you are familiar with the product, it is like a library search system. If you ask for how many "these" are in the file it will come back and tell you how many word "these", and on what documents they are located. We decided to try the product and look at the data stored in the computer center and view it as text. For example we looked at the payroll file with transactions, the master file and salary information as text, rather than as the quantifiable data we thought of in the past. What we did and IBM supported us well on this was in eight weeks put up the IRIS software, loaded a couple of data bases and did some searching. The product is capable of retrieving on any word in the file that you can think of and pull up the documents. You also can do some "anding" and "oring" so you can get combinations of words. What we did was convert information through use of IRIS so that it was retrievable at a terminal. You do not have to have a lot of knowledge like an account number or some special numbering scheme to be able to get at the information. The objective was to make information available to people so that they were location independent and really did not have to receive any reports. The notion is that if I put up every master file in my computer center — that would be all the loans, the savings accounts, the name and address files, the transactions for the last quarter, and all of the transaction files and data files — I no longer needed to print anything. Now, I would like to suggest we have accomplished that. But, the point is we have put up something like the information, master file, transaction files, and history files of every application and formatted it as text so that people who want to respond to individual inquiries can look at information, take some summaries of data, and they can do it from a terminal. An interesting by-product point in putting the information on a terminal, is we no longer have to decide how many copies of a report need to be generated 81

— that is our objective, we are not at that point yet — because if you have to save it once, you have got it for everyone. You have also got the benefit of not only do you have it for everyone, but you have it for the daily user who uses it and really needs it. You also have it for the periodic user who wants to look at it once a month, once a quarter, or only when someone has called him and asked him what is happening. It has turned out to be a good customer service tool because you also have the information available for customers calling in for information. For example, one of the data bases that we orignally didn't even think about putting on was 40 days of checks processed through our bank, something like 30 million items. We decided to put that in an on-line mode to answer customer inquiries. A check which once took 30 to 45 minutes to find, can now be handled in 20 to 30 seconds. It has also eliminated some problems, such as if you called in and called your commercial officer and he called an operating area — and we have got to remember we are playing telephone tag in trying to accomplish these things — and then they called back and said, yes, it has; by the time you have an answer, hours have passed. Since we stored it in the area that deals with the checks, the commercial officer can access the information himself. We are now routing notices with 25 firms saying, do not call us at all, why not dial in and look yourself. So, the whole notion is once you have the information available from and for the source that needs it in the first place, you can now start moving that information, as long as you have proper security in place, to the people that need it or who were interested in the first place. So, we have been converting files of all types. I have my data center performance project information, anything that we need to manage our area we put up under IRIS. We are moving towards — and I don't want to lead you to believe that we have eliminated paper in the organization — we are moving towards people thinking hard about current generations of systems that we are upgrading right now; they are talking about reducing or eliminating a lot of the paper they use to get. To make the full transition of cutting out all paper is very difficult. But, where electronic information really is helpful is where there is a need for distribution of information around the country especially in Europe. These people are responding more quickly because in the past they never got the paper reports until weeks later anyway, and all of a sudden they can have it the same day. So, we are moving more and more toward the information source and retrieval back to the source and having customers actually tap into their own files where they find it appropriate. 82

For people who are familiar with IBM disk, we have converted data so that we now consume something like 90 to l00 IBM 3350 disk drives. What we did do in terms of cost is we installed each application, like the accounts payable system and our other computer systems, in such a way that they would justify the cost of the disk and the incremental cost. The education process and the terminals. We are using Lear Siegler terminals primarily because cheap is good. They cost $700. They are less costly than a typewriter and there are fewer questions regarding the significance of having those on a desk rather than spending $3, $4, or $5,000 on some terminal types. The education process we underestimated significantly. We do have marketeers. We give them territories within our organization. Their objective is to convert more people and convince more people to use electronic media more. We have divided up the bank organization, no differently than IBM and other suppliers divide up their territories, in order to market our product. Our whole goal here is we have a lot of pieces of the puzzle and we are marketing the pieces, not the whole. The whole intention, obviously, is to merge it into a total piece. All of a sudden the manager and professionals will see that they have all of these pieces once they have their terminal installed and are using the information. The area that we now are focusing on and trying to address is, given you have people with location independence and you can retrieve anything you would want about your firm, off the files via a terminal, now you can start offering different communications tools. We all recognize, and I think it was Marshall McLuhan who said, "The medium is the message." We might have some new mediums in which to communicate. We all recognize face-to-face is preferred. We also realize the written message is still useful, and there are times too, when you want to write and when the reader wants to read a personalized letter. The telephone has been around and tends to be used as an interruptive tool. Our focus right now is let us replace the written message where we can by electronic mail and audio mail. We believe that electronic media does it more cheaply and more effectively and provides you more function than you have ever had before. The other area we are starting to test in two ways is teleconferencing. Now, if you have a terminal where you can view all of the information you want to, you might now consider voice conferencing and picture conferencing. In the voice conferencing end, we are now using some picture phones. If you have the terminal dialed up, I can share information with you directly. You can be reading the document no differently than if you were in my office. If you can also see me, what other things will make you comfortable or 83

uncomfortable with the process of saving travel time? We are experimenting with picture phones. The only office the telephone company has in Chicago, is starting to work towards saving some people travel time. We are just trying to find out how people feel. We only have about ten of those installed. Our view is that the real key is the communcations alternatives and starting to think about the time wasted in traveling to and from. In line with some of the notions we are talking about it really opens up some scenarios of work at home. We feel the technology is dropping in cost and all other things are rising very rapidly. It opens up pilots of people working at home, in neighborhood work centers, and just working in your sales territory or wherever you might be most efficient. We have got a lot of feelings on some of those scenarios we plan to test, but that is pretty much where we are today. 84