The Future of Credit Management & Modelling

Bill Weiss:

Well, this is wonderful. I consider this kind of Mount Rushmore of business credit. You know we got some great people and companies on here and it’s really exciting for me to be able to moderate this panel. So let’s jump right in and kind of put everyone on the spot a little bit since they just saw the video but, picking things off the video. What are your views on the future of credit management and I guess we can start with you, Vernon?

Vernon Gerety:

All right, great. I think the thing I’m seeing is the workflow tools that you guys demonstrated here integrating the credit scoring process and the decision process with actually making decisions. That would be the thing I’d see best happening today.

Jan Minniti:

Country. I’ve been doing this for a long time. So it’s interesting to watch how credit has evolved over the past thirty-two years and from just one-man shops to big shops. But we still have one-man shops. So I think it’s amazing how it’s going to expedite people’s workflow. They won’t have to go and delve into stuff that pushes them right away. It’ll be interesting to see how it works with a smaller shop where people don’t necessarily have a lot of people working on a bunch of accounts. They just own a one or two-person shop. That I think there’s a need for that, but this is filling a great need for large corporations.

Bill Weiss:

Absolutely. I definitely want to see this progress. You know to the down market as well. Yes.

Dustin Luther:

Very cool. You know a lot of the points that were made in there would just echo around integrations, you know it being so important and of course. The next couple of us are very much on the data side here and just getting that quality data into these systems is really where I see a lot of this going.

Dan Meder:

Yeah. I think the convergence of data analytics and technology that you saw in the demo was great to see and I think more and more of that, especially with the advances in technology around, you know, machine learning and that sort of thing. Also being from experience, the commercial and consumer data that you were able to bring together. We’ve seen a lot of that over the years. There’s a lot of power in being able to bring the two together and it was good to see that that’s being added because I think there’s a lot there as far as the future goes.

Bill Weiss:

Absolutely.

Ryan Camlin:

Right. Yes. So when I look at this video, I mean what I see, it really allows the human to focus on the critical thinking aspect versus all the other manual tasks, right. So I even look at this specific use case as being very much around the manual kind of report reading piece which I could see that becoming less and all data is brought in and more of this is automated, but it really makes it very intuitive to actually do the manual report reading. I think that’s going to make that process a lot more efficient.

Bill Weiss:

Great point, great point. All right well let’s get to the second question. Credit Data is one of the most important factors that determine automation. What steps are the credit bureaus taking to improve access to data? In fact, Ryan let me start with you.

Ryan Camlin:

Yes. So actually when I think about this question, I really think about the things I’ve done in Bradstreet, doing especially around our partnership. So we partner with the Small Business Administration to help the under banks and the underserved to gain access to credit. We’re partnering with the State Department on the human trafficking index. And so that’s going to help companies identify the real risk of modern slavery within their supply chain. And we partner with the Virtue Foundation as well and so what they’re doing is they’re trying to find out what are the most efficient ways to dole out aid money around the world. And they’re using our data to help enable that process.

Dan Meder:

From our perspective, we’re doing a lot of work around making our data easier to access. We’ve launched an API hub, where you can come in and make one request of Experian and have access to all the data assets that Experian has across not just commercial but automotive consumer and all that. So making it easier to work with the data and also encouraging developers, not just our clients but also developers to come in and kind of kick the tires because there may be folks that are developing applications for you that would like to understand a little bit more on how they could work with the data. So we have a developer site that we allow people to come into and kind of like a sandbox kind of come in and play around with. So we’re also starting to bring through third party data as well. We just launched a protocol hazard hub which is insurance data. That’s being brought in.

Dustin Luther:

One of the things we’ve noticed you know, credit games started in Europe and there’s just a lot better like small and medium-sized data there. So one of the things that we noticed in the U.S. is just that data is pretty weak in general across everybody, it’s a pain point for us here. So a lot of our emphasis over the past couple of years has banned getting better data on those kinds of smaller and medium-sized companies. So, just in the past year, we have three different products a financial trade product that gives bank data banking data and we have this my credit save, so we can kind of consider it like a small business trade tape where small businesses can bring their information in automatically and using business principle reports similar to what you’re showing there. Just one click, you can come in and run the personal credit check on smaller businesses. So those kinds of things I think is where you’re seeing us get better at doing our best to get better data into this system.

Bill Weiss:

Excellent, excellent.

Jan Minniti:

It’s hard to go after everybody has said everything. I would agree. I think all the bureaus we represent, basically, all the different bureaus have really worked hard to bring all different kinds of data to you from trade data to public record to news to everything. It’s so much easier now where everything is coming to you and pushed to you. And with this video, it even pushes more to you to say it because you can ask it to go and grab data that you need when you need it. The fact that you can pull on the owners, the businesses that all public record everything. I think it just makes credit manager’s job so much easier because all the data out there now is so much more visible and the bureau is going out and searching and great gathering it for you, where you don’t actually have to go out and gather as much you can go to one-stop partitions to stop and get everything.

Bill Weiss:

It’s great to have that perspective since you work with all the bureaus.

Jan Minniti:

Yeah, it’s just that they’re doing a great job of getting information for everybody to make your lives a lot easier.

Bill Weiss:

Great. Thank you. All right. Next question. And we can start with Vernard on this one. I think one is good for you. How if credit models evolved over time? You’ve been a few of those.

Vernon Gerety:

Yeah, I’ve done a few. I think the techniques, they’ve been there and they’ve been sophisticated. I don’t think everybody is quite adapted to them. A lot of people would go, I’m doing a session later today about an expert scorecard which worked quite well. But I think people are beginning to adopt the statistical base models and to do all the pounds point. The other big thing is that data’s getting so much better. So it’s kind of garbage in garbage out. So we don’t have good data models, no matter how great the model is, it’s not going to be predictive. Now, what I did like about the video is, I don’t really see AI at this moment at least especially on the smaller commercial space. Actually being too integrated into the credit piece in terms of the credit decision. But I like the idea where it takes the rating and then turns it into a logic flow in terms of what needs to be done next. But the other thing is, of course, the person.

Jan Minniti:

Scoring models, so when they started there was just an initial pay score model that told you the probability of being paid late or now you can get there so many different models. You have the probability of the company going out of business now versus the probability of late payments. Two totally different things mean different things for each one of you. There are models based on large companies and small companies. So over my career, I’ve seen the increase of just a basic model that was meant to fit everybody to new models that actually fit the size of the company you are selling to, the size of a company you are, your risk, and things like that. I think the models, all the different statistical models have really been, the bureau’s been really good trying to create something that fits every one of you. And of course, then you can always go ahead and have a model created specifically for you. But I think that the growth in just the past 20 years has been exponential. Just huge.

Dustin Luther:

I was actually going to make a very similar point which is these very specialized models. You’re seeing people grab our data and grab other data sources and do really interesting, very often industry-specific like you say that’s only another piece I would add would be the industry-specific models you’re seeing come out which is quite interesting.

Dan Meder:

Yes, I was fortunate to be one of the product managers when D&B first launched credit scoring back around ‘89 or ‘90, somewhere around there. And have had the opportunity to kind of see the changes throughout the majority, almost 30 years I guess. Vernon made a point about judgemental scorecards and your risk models which everybody was very comfortable with back in the early days. You know, because you could see inside the black box and you can understand how the calculations happened and that was very important to everybody. But over the years, I think everybody to Vernon’s points got more and more comfortable with statistical models. And now we’re starting to move, it seems like into another generation where it isn’t just how do we? But moving on to things like using statistics to assign credit line increases. Looking at statistics for account management purposes, you know, you saw a great example of it here. I think anyway if I was reading the tea leaves the right wayside video around using it for collection strategies and things like that. It experienced and again kind of to Vernon’s point about the use of A.I. and machine learning.

But our consumer business is using it a lot on the fraud detection side and for a very good reason. Fraudsters move at a rapid pace right. Once you caught onto him they’ve evolved into another way of trying to scan your system. And traditional models can’t keep up with that. So we start to invoke machine learning which can then take all these inputs and start to really derive patterns very quickly and see and evolve kind of with the fraudsters and try to stay as close in lockstep as possible. Now, the way that kind of translates into a use case at least on the consumer side that it has so far is that you know you get a lot of false positives if you do it wrong and you don’t want to accuse somebody of being a fraud if they’re not.

So you tend to bump those out from manual review which kind of defeats the purpose of an automated decision process. But with machine learning what we’ve been able to do, we have one use case that has some very significant results where we were able to reduce the manual reviews by like 74 percent because we were doing a better job of targeting the actual frauds and not so much trapping a bunch of non-fraudulent businesses in the same net. So that’s, I think is a great opportunity for everybody going forward as that technology evolves and you know it can really put a finer point on the productiveness of the scores.

Ryan Camlin:

Yeah, I mean, so I think firstly we see a lot of companies therein growth mode. So they want to be able to extend credit to as many people as possible within reason. Right. So I think what we’re seeing in models is the use of more alternate data so they can automate a lot of that within the models. I think the other things that we’re seeing is really the use of multiple models, where we used to see a single model and especially field testing multiple models against each other, sometimes for a matter of months.

And it’s not just really understanding which model performs the best but also which one is the most explainable as well. So it’s the kind of combination of those two things. I think you know for the most part we’re still seeing the regression models especially for origination, but I think more and more you know, in the last couple of months and really last year we’re starting to see requests now for machine learning-based models as well even for origination on the commercial side. Right. Thank you.

Bill Weiss:

All right. The next question, credit managers in financial services have always been at the forefront of innovation. What are some ways in which corporate credit managers could get inspired by the financial services sector? Ryan may be to start with you.

Ryan Camlin:

Yeah. So I think you made the point about fraud and I think when I think about financial services there- for one, they’re in a highly regulated industry. I think fraud is a big problem there. And so, I think we have had to be at the forefront of things like machine learning to be able to identify those patterns of fraud and then they’re using you know RPI cognition, they’re doing this to really enhance their KYC programs enhance their compliance. And really, even on the consumer side you kind of look at the self-service and these touch lists and frictionless experiences and I think you’re going to start to see that really even on the corporate side as well.

Dan Meder:

Yeah, I think those are all great points Ryan made. Because we’re seeing very similar things from our side. I think if you also take a look at the history of how financial services organizations have had to really rely very heavily on analytics to manage their businesses and a lot of it’s been due to just the sheer volume of other counts that they get. They had to offload a lot of that onto the models. You know, you’re not going to hire enough people fast enough to keep up with it. But also what it does is it gives you some real objective measurements to determine how to take action.

So credit approvals, that’s kind of an easy one. But also, if you start to look at the behavioral models to look at account management, you saw here I think a good example of collection strategies being used or being driven by statistical models. We’re also seeing the use of it on the data management side as well. So, what this all means is as you evolve more towards a frictionless type of environment across the whole lifecycle right from credit approval all the way to the collection. If you’re going to move into that you really need to pay a lot of attention to the science behind the data evaluation because you need that feedback loop right.

Because you’re not going to have the people to necessarily sit on top of it and you probably don’t want to either. You know you want to have it. I think Ryan, you made the point we’re talking about before. It’s almost becoming more of a customer service kind of application and you don’t have analytically trained necessarily people that are going to be looking at these things. So you need those models in the background giving you the right information. So you know what to do with a particular account.

Dustin Luther:

It’s interesting because it’s fun to be in the middle. Sometimes you guys are saying it like actually I’d know what I kind of repeated. I think I’ll just let you go. Dan, I know I was the next, you go on.

Vernon Gerety:

I got to agree with everything Dan said. No, the financial services space that’s pretty much my biggest segment in terms of building models. They were early adopters and came over from the consumer side and they’re completely into automation. And the mantra that they have is those with the best risk strategy win. Because they’re effectively building a portfolio of assets and the better quality those assets, the better their profitability is going to be. So I think the automation and the ability to make a decision is very key for them.

Bill Weiss:

Absolutely. Thank you. All right. Next question. Credit Management as a profession could get oversimplified to geeks applying some mathematics to arrive at a decision. How can credit managers play a more strategic role in the organization? And since I said math and geeks in the same sentence, I think you really should go first on this one.

Vernon Gerety:

And actually, I make a bet I’ve been making a very strong point for a long time and I’m actually going to talk about this in my session. The whole idea of using a credit scoring model effectively is to make sure it falls within your risk philosophies, your risk policies, and your risk standards. So when I work with clients and I throw a model too much and I’ve just thrown on the wall you basically say we’re going to work with you to make sure that models getting you to the end node that’s giving you the right decision that you’re comfortable with. And you hear a parallel process in all those sorts of things. But yeah, this seems like this is a good score. But let me tell you the three reasons that there is a problem with that. And if you see enough, but that is a pattern you say that we’ve got to go back and fix it. So I think that the thing I do with my client’s work with their risk philosophy to get a model is a kind of they’re going to be comfortable with.

Jan Minniti:

OK, can you repeat the question?

Bill Weiss:

Sure. Credit Management as a profession could get oversimplified to geeks applying some mathematics to arrive at a decision. How can credit managers play a more strategic role in the organization?

Jan Minniti:

OK. Actually, credit managers have to play a more strategic role in the whole aspect of the business at this point. What you’re seeing is and I hear it all the time is a lot of credit is not the shiny part of the company. They kind of get overlooked and think that people need to get out the credit managers, credit departments. If you don’t get paid, you don’t make any money. Although sales are great. It’s a gift if nobody ever pays it. Credit managers aren’t looked at as highly as I feel they should be because you guys have really important roles. And so I think if you can show your value through education and through statistics doing modern reports and stuff and kind of push it upwards to the C Suite. So they see your value that I think that you’ll have a lot more input into where the company is going, what’s the future for the company and they’ll want your ideas, they want you to be involved and I think with especially looking at automation, I think it’s important that credit managers remain really relevant because you don’t want to automate. I mean there are always going to need a credit manager but you know you want to make sure you’re relevant and with everything that the future holds.

Bill Weiss:

You guys want to make their jobs better.

Jan Minniti:

You do. You wanna make their jobs better. Now that sometimes people won’t go up to the higher-ups and talk to them unless there’s a problem. You solve tons of issues that aren’t necessarily problems you go and say- look this and this you’re here at this educational format you can go and say look what we did and look what we found out that we can offer to the company to save money, make money do all that and you’re just so relevant and so make sure you let your higher-ups know that.

Dustin Luther:

Yes, you know along those lines like just being an advocate for growth is well within the organization. It’s a great spot for a credit manager. So, you know, rather than being really like you envision yourself out there. How do I, as in my role, help these other organizations you know, help our company continue to kind of solve problems, if maybe there are ways to companies you want to work with but aren’t meeting scores, aren’t meeting you know thresholds and how do you kind of get them into the fold in a meaningful way. If it makes sense.

Dan Meder:

Yes. I think my thoughts would be a little bit more generic and that is don’t shrink from the change, lead the change. You know you’ve got a great opportunity here. I’ve had a chance to look at some of the sessions and some really great sessions are going on at this conference. And you know, learning about what’s possible and taking that back and getting out in front of it and being the early adopter. So to speak, you know or leading the change. One word of caution of course in all of these things is you want to avoid getting seduced by the shiny object, right. I was at a conference two weeks ago, it wasn’t credit-related, it was asset management and they started talking about how two or three years ago there were a lot of guys who were pitching alternative data satellite imagery and all this kind of stuff. Two or three years later they’re not that interested. The asset management community isn’t as interested because as they kick the tires on it they’ve started to find out that it isn’t very deep and it really isn’t that useful. So it’s important to get out in front of the change. Learn as much as you can when you’re in venues like this because it doesn’t get any better than this. But then also be wary of the shiny object because the last thing you want to do is go in and say hey-I think we ought to go in this direction and find out. You kind of missed the mark. So anyway, like I said, fairly generic advice I suppose you could apply it across a lot of things. But I think it certainly would apply in a situation like where you are today.

Ryan Camlin:

And I kind of build on both points, I mean the point about growth and the point about driving change. So when I think of kind of a credit manager being a catalyst for growth it’s really making sure that your credit policy is aligned with the corporate strategy. For growth, what is the right risk tolerance? Now I think the other piece is really being an agent of change in the organization. So driving digital transformation within the organization. I think you know the credit department finance operations can really be that change on the management level. And I think you know the third piece is really the credit manager. I think he really understands the customer, kind of owns that record. They know you know who’s paying, how much are they paying, how often do they pay, really kind of giving that 360-degree view of the customer and sharing that back out with sales and marketing to help drive growth as well? Thank you.

Bill Weiss:

Well, let’s take a pause now and see if we have any questions from the audience. I have a couple more questions that I can ask. But I wanted to make sure we had enough time to field any questions from the audience. Hopefully, it’s not too early. The question right here.

Audience:

Good morning. My name is Doug Dunlap. I’m the director of credit at Tetra technologies. And my question is all-important, what we’re really ultimately responsible for is the cash flow of the company. So if we’re doing our jobs right and everything’s working, we are taking some risk, we have a percent of our portfolio that are labeled as high risk. We treat them differently. We understand who they are but are we taken enough. Is there anything that you have thought about or that could be in the future that would actually take a look at my portfolio and how we’ve extended credit, how we’ve recognized dome as a risk grade and help me understand. Are we taking enough risk? I’m always afraid we’re leaving money on the table. At the end of the day, every sale that comes to my department the answer is yes. We just apply a little bit different tactics on how we’re going to define that relationship. But every sale is yes. Some of us cash upfront but the sale still is yes. But I often wonder if we’re leaving money on the table by not taking more risk. And I would very much like to have your opinion looking at my portfolios or anything like that.

Dan Meder:

Just real quick. First of all, if you’re looking at your specific portfolio there are custom models that can be developed around your specific portfolio. So, it reflects. Not a generic sort of general population view of the world but specifically your book of business. And what comes out of that is a list of probabilities. So that is very objectively derived. So, for example, a score of let’s say 1 on a scale of 1 to a 100 would be really bad and it might come out that you have a 75 percent likelihood that the business is going to go bad. So you might want to stay away from that. But if you get to- let’s say, a score of 20. And the probability is like 5 percent or 8 percent or something like that you might then say, hey listen in the past we might have turned that down, for this reason, or this reason but because of the way that scores sit relative to the way the performance in my portfolio has been. I might take that risk and run with it that way. I mean you’ve made a living doing this stuff.

Vernon Gerety:

Yeah, I can do that actually. I guess it goes back to my earlier point. Those with the best risk strategy when so you have to understand the risk you’re taking and you made the point like pricing for risk effectively you’re pricing for risk if you’re shutting downturns or cutting down credit lines. I mean the financial services that could actually change the rate they charge, so they’re much more proactive in it. But yeah, that’s something that can be done and it’s highly recommended that you understand every account you take to dance point. What’s the real risk? What’s the probability default? What’s the probability delinquencies?

Ryan Camlin:

I would answer that just a little bit differently. I mean, I think for one thing at the portfolio level is the right way to think. But in some ways, I think we, as an industry, probably haven’t gone far enough in terms of the kind of analytics we deliver. We have descriptive analytics. We have predictive analytics. I think the next step is really getting and fine-tuning really prescriptive analytics, where it’s telling you right, it’s saying these are the steps that you need to take. Right. Having that very clear recommendation as to the output of the model, I think is the next step and I think it’s going to make you a lot more efficient and really start to unlock cash flow.

Vernon Gerety:

Speak for yourself.

Audience:

Hi, I’m Don Howard on the area credit manager for builders’ first source here in Houston and this question is specifically for Jan. I participate in NACM locally. So my information is pulled from NACM Southwest versus there are also other reporting areas for NACM across the country. I tend to see that the information is different when it’s pulled from different reporting areas and I’m just curious if NACM as a whole has a plan to bring all that information together.

Jan Minniti:

OK. So if I understand you, you’re thinking that certain area parts of the country have more data than other parts of the country. It’s really the different individuals NACMs have the option to participate in the national trade report. For instance, our affiliate does and we’re out of the Seattle Western Washington Alaska Hawaii and all that. So it’s really up to the individual NACMs. I think that as you’re seeing the changes NACM and there have been a lot lately you’re going to see more growth in that arena where there’s fewer NACMs, more larger affiliates and the larger affiliates will more than likely contribute all their data to the national trade reports and then have a big growth initiative. I can’t really speak too much on it but I think that you’re going to see in the future a lot more growth in the arena all across the United States. It is an option just that’s the thing you have to look at. So it’s up to the individual affiliates what they contribute or not contribute to the database.

Bill Weiss:

Thank you. Other questions.

Jan Minniti:

Somebody back there.

Audience:

Hi, Nick clone from Apple. So just a quick question, as you think about the future of credit management and modeling and you think about areas with the highest ROI so return on investment. What areas are the most attractive that you’re looking at towards the future and that offer some of the highest returns?

Bill Weiss:

I’m not sure I totally got that.

Audience:

So if you think about the future of credit where companies could invest to improve their capabilities to model say statistical modeling or their ability to manage credit. What projects do you see, what areas do you see that offer the most compelling returns on investment? So if we’re looking at a capital budget and capital outlays for these types of projects, where do you see them being most enticing for the return in some of these future areas of development?

Vernon Gerety:

I’ll start, I think the band’s point earlier I think that behavioral modeling, especially in the trade credit world is probably your best focus onto the gentlemen’s question earlier in terms of understanding the risk across your entire customer base and the fact that you build these relationships and then once you build a relationship you get a lot of repeat sales. It’s very important to see what’s dynamically happening within your customer base. I’d say looking at your customer base on not just a monthly basis but basically every time something changes we evaluate risk with that customer.

Dan Meder:

Yeah. To build on what Vernon said, behavioral models typically rely very heavily on your information. So it’s taking into account how you’re being paid. I guess, Vernon, you get it when usually when you build it you do it over time. It isn’t just, it would be true behavioral models there’s a lot of trended sort of impact on how you evaluate.

Bill Weiss:

Anyone else or we get with that. Hope that answers your question. We do have time for another question. Anyone have any questions. One more.

Jan Minniti:

There are so many back there.

Bill Weiss:

Okay. Yep. Sorry.

Audience:

Good morning. My name is Belero Alvarez. I work for chemicals. I have one question about D&B because usually, every time I try to find something, there I find the same company several times with different numbers and information in each one is totally different. I want to know why you are unable to fix that or get me any information that will be useful for me because when I had to go to different accounts. Well, we’ve put in the numbers, the information that I said is totally different.

Ryan Camlin:

Yeah, I think you know it really depends on the entity. So I mean that’s very specific examples or specific industry that you’re in but I think it’s always a challenge in terms of, you know, you’re collecting all this data you’re linking it to a single entity. You know those companies may have different relationships. They may be viewed as different entities by D&B. So we may be treating them that way. So it’s really kind of making a decision for understanding what’s the true legal business entity then associating all the data back to that. So it may seem from a certain perspective that you know these are the same business right but they may actually be different legal entities. So that could be driving some of that, but I certainly think that there’s, you know opportunities for improvement in terms of how we can make that more intuitive through search and through some of the tools that we have that are out there.

Vernon Gerety:

And I would just generally say that fragmentation of data is prevalent across all databases. We don’t have a Social Security number of a business that you get when you’re born as much as easy to track. So that’s kind of an issue for all these databases.

Bill Weiss:

One more question over here.

Jan Minniti:

Gunrunners get their exercise. I could do that.

Audience:

I’m starting to realize the amount of sort of people aspect involved in the credit world here. There were a lot of key buzzwords thrown out during this conversation about relationships and people’s behavior. My question is, from the talent side. Right from, you know when you’re looking for people to help you do your job. You know how much of the people aspect and the people skills are you taking into account. I’ve realized this is a data-heavy space but it feels like it’s also a very people-centric space as well.

Ryan Camlin:

I think if, for one, it depends on the role. So if you’re thinking from the perspective of the credit department, right if that’s what you’re questions related, when I look at stuff like this the importance of having someone in that role that has people skills and that is customer-centric is going to get just even more important. Right, as more of these tasks become automated. If you’re thinking from a bureau side, I think it really depends on the kind of role that someone is playing. But I think it just in general as we sort of move through the knowledge economy those types of people skills are going to become more and more important. Overall.

Dan Meder:

I totally agree with everything Ryan just said on the customer service side and having those people skills. I think, Jan, I kind of queue you up on this one because you had brought this up when we were talking before- Is that you still need people who understand the analytics and still understand the process that is used to create those analytics because if you lose that experience, your models are going to take over or they’re going to decay or something like that. So you know and I think you made a good point we were talking before – if you move almost too much towards a model that says OK, we’ve become more and more customer service and all the thinking is being turned over to you know the machine or whatever it is, you’re going to lose a lot. And you know, you’re not going to be able to help evolve that.

Jan Minniti:

So yeah we were talking about this morning and can you hear me? So my concern is again having done this forever, it’s different and I actually believe this is a great product and all of that goes along with it and I’m all in technology. My concern is if we don’t pick good analysts. If you don’t pick good analysts and you may not need as many but here you’re going to have to really pick people who are really good because someday they’re going to have to replace you. So if you’re picking people that just can read or are listening to what a machine is telling them to do but don’t have the ability to think beyond that then, who fills your shoes when you move on and then what happens to the credit department? You know, you’re always going to need somebody to tell the machine what to do. The machine is never gonna be able to totally figure when the economy changes, you’re going to change things, you guys will have to tweak the system so it works to your ability, where you’re going to need somebody to follow up behind you should you ever want to a move to a different position. Different jobs, retire things like that. So I think we need to somehow figure out how to grow the people that are coming in and keep their interests so that they’re not just sitting behind a machine doing what it says. You have to give them something to grow them so that they stay. They want to continue that path because most credit people started as analysts, most of you didn’t walk into a credit position at a college. You started somewhere and grew into who you are today. So that was more what I was talking about earlier.

Bill Weiss:

Thanks, Jan and thank you very much panel. We are out of time but I highly recommend it if you want to learn more about credit. Vernon has a session later today. He is a college professor in his spare time as well. You’ll definitely learn a lot and Dan is also having a session later this afternoon about 3:15 again. Highly recommend that you attend those sessions. Thank you very much.