Fraud Prevention & Detection: The Case for Data Analytics



good afternoon and welcome to today's webinar this is the first in our four part series on transportation and analytics today's webinar is going to be focused on fraud prevention and detection the case for analytics and what we're going to do today is talk about how analytics can be used in your organization specifically in the transportation industry to help you prevent fraud and also detect frauds that may or may not have already occurred in your organization the rest of the webinar series is going to focus on vehicle analytics driver based analytics and then also how you can incorporate analytics within the internal audit practice within your organization so I think it's going to be a very fun series we've got guest speakers coming in and experts in each of the areas joining us for this webinar series I think it should be a lot of fun and with that we're gonna go ahead and get started so to start with I want to talk about why we need to focus on data analytics and I really like this quote by Eric brillo and this was given in his TED talk and it talks about why we really need to step back and embrace complexity and really when it comes to analytics I think this quote couldn't be any more true it says that we're discovering in nature that simplicity often lies on the other side of complexity so for any problem the world we can zoom out and embrace that complexity the better chance we have of zooming in on the simple details that matter the most and I don't think that that can categorise analytics any better really what we're doing in data analytics is we are zooming out rather than focusing on a subset of data maybe a sample of 30 or 40 or 50 instead we're focusing on the entire population of data we're leading looking at a hundred percent maybe that's a million records maybe that's five million records maybe it's just 50 thousand but really we're looking at everything and once we look at everything and apply a certain set of analytics certain tests or procedures on those data sets now we can see where the high risk factors are and then we can focus in on those details that matter the most to us and really research the various elements transactions attributes that we need to that may be the most high risk so I think that we really need to take this quote and use it as the foundation for this entire webinar series we're going to zoom out we're going to look at something that's large complex large data sets a lot of data sets of a variety but we're gonna do that so we can focus in on really what matters the most and get down to that granular detail later on down the road so today's webinar just a quick road map as to where we're gonna go we're gonna start off by setting a little bit of a foundation as to why it's so important to talk about analytics we're gonna try to break down some barriers and get everybody on board that this is something that you can do after that we're going to talk briefly about some trends in fraud detection and controls that the Association of Certified Fraud examiner's has out just to kind of set the table a little bit there then we're gonna bring in an expert here from BK DS forensic evaluation Services Division Jeff Roberts he's a managing director and get his insights on just how data has changed the landscape of investigations and how investigations really are being conducted within the transportation industry after that we'll come back for a few closing comments and then we'll definitely take questions from the audience now we're gonna go ahead and start to set the foundation a lot of times when I work with organizations I found that really the challenging getting started with data analytics is fear of data analytics and the fear of big data and big data is something that we hear about so much in to the world today whether it's in the media whether it's in advertising whether it's in the news or sports even we hear a lot about big data and if you were to google the definition of big data you're gonna find a lot of different definitions but I like to go to the source which is gardener and the gardener group they specialize in technology and data and I really think they're a great authoritative source here and their definition of big data is information of extreme size diversity and complexity so really what they're saying is big data is data that's big and I would like to think that pretty well everybody can get on board and understand big data given this definition and really it doesn't matter the size of your data now in the transportation industry there's an enormous amount of data out there and I would venture a guess that pretty well everybody watching this webinar has big data but even if you don't see that you have big data maybe you think you have medium data or small data or you have no idea what size of data you have it's really irrelevant its data and the principles and the methods that we use to analyze it really are largely the same and those methods and procedures that we're using are really the definition of what data analytics is data analytics is really it's a set of processes and methods and procedures that we're using to extract the most useful information out of our data and answer strategic questions and that's really where it starts with data analytics is asking a strategic question the one we're gonna focus in on today is where in our organization does fraud occur where has it occurred where might it occur but that strategic question is what's going to guide us on how we apply data analytics within the organization now that strategic question is going to change in the other webinars in this series but today we're going to focus on where in my organization is fraud occurring now analytics is very beneficial to our organization and this is why we really need to focus in on it on the screen you're going to see responsiveness continuum and this is really how do we respond to our strategic question and using what I'll call the old-school method of going to the boxes and pulling a sample of 30 and digging through the paper that's very reactive we're not gonna do that and actually have a good chance of finding fraud in fact we have a very low likelihood finding fraud using this method so we want to get away from this reactive sampling paper-based approach and we want to move into a much more proactive frame of mind and data analytics is going to help us to get us there data analytics is gonna provide us a hundred percent coverage we're gonna be looking at every element of the data set and we're gonna be applying a set of rules or maybe indicators maybe behavioral based tests to that data but it's going to be on a hundred percent scale so we get a lot more coverage now to be truly proactive we need to move into what's called continuous analytics and this is where we have routines and procedures that are set up to run almost all the time now it doesn't have to mean real-time where it's every second of every day this could be at the speed of your business process a great example here would be purchasing card or fuel card data if you wanted to analyze that at the speed of business you're probably going to need to get that data on a daily basis analyze it every evening have it set to run overnight and then you can follow up on exceptions the next morning that's continuous analytics and that's truly being proactive and getting out there trying to get ahead of perhaps the fraud that could be occurred in your organization now as we go through each of these webinars and especially starting with today's I want to provide everybody an application framework so how you're going to be able to think about the procedures and what we're doing and how it's going to help you implement analytics in your organization and it's really a six phase approach the first thing you want to do is assess your risk now this is obviously on the fraud detection side is where in my organization do I have risk there may be multiple answers to this question but you're going to focus in on just one of those in Phase one once you have identified your risk you now need to figure out what objectives do you need to meet to better manage this risk if we're talking about looking for employee vendor matches your objectives might be identifying employees that share a social security number with a vendor maybe an address with a vendor the objectives are going to vary by risk but you want to come up with very specific objectives in phase two after that now it's time to gather the data now ideally you've already thought about data as you've gone through phases 1 & 2 and you have a pretty good feel for what you need but in gathering the data you want to make sure you get two things the first is what you're going to be analyzing the second is what you're going to need for follow-up and research so in phase 3 we obtain all of the data that we might need to analyze or follow up and then in phase 4 we get to start developing procedures and in my opinion this is one of the most fun phases what I would like to caution you on is don't try to come up with the most complex procedure right off the bat what I would recommend is taking the procedures and breaking it down to the very basic level take the easiest 1 step procedure create that first then create the next step and the next and the next and so on after you've created each of these steps individually then you can bring them all together into a much more complex procedure maybe you already have those and now you're ready to automate those procedures but it's definitely a progression that you want to work your way through you don't want to start off with the hardest most complex thing right off the bat once you've developed the procedures are going to apply them and now you have a set of results and now this is something that we're all familiar with but we have to get better at and that's analyzing the results what exactly are these results telling us do they tell us we have a lot of false positives and our procedures aren't very good if that's the case we need to go back and change the procedures to do our results tell us that we have more risk than we initially thought if so we may need to revise our objectives but we definitely want to make sure that our results are helping us to meet our objectives if they aren't we need to go back and we need to develop better procedures so we get better results the final phase is now managing the results and a lot of times this means pushing out what you have learned to the various business processes within your organization so if we've learned that we do have some risk and employee vendor matches and a lot of that is due to the vendor setup we need to go back to our accounts payable in purchasing department and educate them on the risks that we've seen and what they can do to help us better manage those risks then we can develop procedures to see if that's truly going on but we have to use these results and manage these results so we can truly better mitigate and better address the risks that we identified back in phase one so as we go throughout the program today think through this framework when we're talking about the various investigations and techniques that you may have out there the various data sets we may use and think about them through the context of this framework going through where they might fit in phase one two three four five or six the ACF he produces a report every couple years called the report to the nation's and what I like about that report is it does tell us how frauds are detected and it also talks a lot about our anti-fraud controls now there's a lot of other information in that report that I encourage you to read and we'll talk about later in this program but we're going to focus first on how frauds are detected and as you can see on the screen there tips are number one always have been my guess is they very likely will be for years to come but the second and third items on this list of how frauds are detected are management review and internal audit and both of those involve a heavy element of data analytics a lot of internal audit departments I talked with they're using data analytic and they're using it for fraud detection a lot of management review is based on data sets that they're reviewing so while it may not be something you think of right off the bat when you think of your management review you're likely using data and data analytics to a degree in doing that now the reason that I show this slide is actually the fourth item on here which is my biggest concern and that is the fact that the number four way that frauds are detected are by accident and this has got to be one of the worst controls that you can have built in into an internal control framework and that's we're gonna catch fraud by accident good luck right there's really not a good way to build by accident into our framework so my hope is that more organizations can start to use data analytics and by doing so will continue to take this number down to where it ranks at the very bottom of how frauds are detected and it's no longer up there in the top four or five now new this year in the anti-fraud control category proactive data monitoring and analysis and this is something that I'm very glad has finally made this report and the sad part is here when we look at how many organizations are using it we can see that it's just over a third so barely more than a third of organizations had proactive data monitoring and analysis that were part of this report now the companies that are part of this report did have a fraud so I'm hopeful that all of the other companies out there that didn't have a fraud it's because they were using proactive data monitoring and analysis may or may not be the case but I can always hope now what concerns me here is only a third of the organizations are using it but let's look at the benefit that we can have from using data monitoring analysis in our organization this shows median loss and how its impacted by any attack on trol and we can see that proactive data monitoring and analysis is number one ranks at the top of the list first year in the survey and it ranks at the very top and you can see that while it's only used in a third of cases it really resulted in a median loss reduction of nearly 60 percent so nearly 60 percent of a reduction in your median loss by using proactive data monitoring and analysis now I'm not gonna jump to that conclusion real fast and say it's by far the best anti-fraud control out there I may or may not believe that it's one of the best but let's go to the next slide which also talks a little bit about median duration and this is how long a fraud lasts and we can see that using proactive data monitoring and analysis actually has a 50% reduction in median duration as well so it ranks number 1 on reducing median loss and it ranks number 1 tied for first I should say on reducing median duration I think right off the bat we can all agree proactive data monitoring and analysis is definitely effective it's something that we should be using now I will say that everybody that's watching this webinar as long as you have Microsoft Excel you can be doing some sort of data monitoring and analysis some sort of data analytics using Microsoft Excel I know I had a case last year where I was a over a 1 million dollar fraud and if somebody would have performed a simple test in Microsoft Excel it would have been caught the first check that was issued not saying you have to but I'm saying it is very effective and here's why it's effective an example to really drive this point home let's take an example of a picture on your screen you're gonna see a picture and for this picture I have provided you a 10% sample and I would argue that nobody on this webinar has ever pulled a 10% sample of their transactions if we think through that on a million transactions that means you're looking at a hundred thousand of them not likely to happen most people pull 3050 maybe a hundred but here is a 10% sample of this picture and your goal is to tell me exactly what this picture is every detail in that picture that's essentially what you're doing if you're pulling a small sample and trying to find fraud in your organization it's not very effective because as we look at this picture we can see that it's actually a very beautiful countryside a nice bright blue sky like clouds a few date a few silos a barn a tractor a lot of other details that in a 10% sample we didn't have a clue were there and really this is representative of what your data landscape looks like in your organization you may have data in silos it may be segmented across your organization it may all be in the barn who knows you may have all of your data in one big barn or data warehouse and that's great but until we look at the big picture we look at every element of this again zooming out to that complex level and looking at a hundred percent of our data we really have no idea what our data is telling us so this is the example that I always use to drive home the point that we really need to be using data analytics if we want 100% coverage and we want a true picture as to what's going on in our organization we're gonna move into our Q&A session and we're gonna have Jeff Roberts who's a managing director and be Katie's forensic evaluation Services Division join us and Jeff brings a couple decades of experience almost to us today he's worked in a lot of industries he's done a lot of fraud investigations and is one of the best when it comes to fraud prevention and internal control so with that we're gonna welcome Jeff to the webinar today well Jeff thanks for joining us for the webinar really appreciated having your expertise as it relates to fraud investigations and prevention what I wanted to talk about to start with it's kind of starting out a little bit more broad necessarily just in the transportation industry because I think there's a lot that can be learned from other industries and just general fraud trends overall so could you talk to us a little bit about what are some of the current trends that you're seeing in fraud I know the Association of Certified Fraud examiner's has their report to the nation's that comes out every couple years what are we seeing what are we learning out there well fraud continues to be a really really big problem the average company loses about 5% of its revenues every year to fraud waste and abuse that's according to the Association of Certified Fraud examiner's a survey that you mentioned that came out in 2014 and that stayed pretty consistent over the last few surveys the big thing on our end that we're seeing is how long some of the fraud schemes are lasting a lot of people who I talk to think well you know a fraud may last six months three months it's not going to be that big of a deal we can find it and deal with it it's just a part of cost of doing business but we've seen frauds go on for 10 years 15 years we had one just a few years ago that went on for 20 years so these are devastating to some companies especially the smaller and medium-sized companies it cost them sometimes millions of dollars Wow do you see a correlation between fraud you know how long they last and the losses and the recovery what do you see along there the length of time that a frog goes on the greater the losses and in the smaller companies is where we and the medium-sized companies where we've seen the longest frauds go on the bigger companies tend to have the better controls and so typically the losses are a little bit less there but still it's something that most companies need to be aware of okay I'm guessing a lot of companies probably aren't recovering a lot of these frauds you're not running into people that are just squirreling the money away for a variety right I think the statistic according the ACF E is about 60 percent of companies recover absolutely no along those lines insurance tends to be the one way that companies are able to recover fidelity bond crime coverage employee dishonesty policies or the way that a lot of times the losses are recovered all right so it sounds like fraud is definitely something that's here and sounds sadly like it's here to stay which I don't think it's probably any shock to anybody yeah as far as it relates to the transportation industry specifically I know the AC Fe has different rankings and different tables in their report how exactly does the transportation industry rank on that in that report yeah and I think we have a graphic that we can share with the audience on this one okay the transportation industry ranks about in terms of at least the median loss about nine or ten in terms of all industries out there so banking health care manufacturing it's about $200,000 I believe is the median loss per fraud scheme for the transportation industry so it's nothing to to sniff at and say well it's not that big of a deal because again as I mentioned earlier you get these smaller and medium-sized companies they have a really hard time recovering from a loss they may be just getting by and when they learn that they that they lost that much money not through their operations and maybe some decisions that weren't quite right but rather through an employee it's just it's just is so devastating right that two hundred thousand makes a big difference it does if it's maybe audit materiality which a lot of people in finance and accounting fear about it may not be mature my guess is most the organizations that you've run into if somebody steals a couple hundred grand there they're concerned absolutely their concern yes as far as the types of schemes that are victimizing transportation and the industry itself you know it's good that we're not talking about mining or real estate where they're losing nearly a million dollars right so I'm guessing that the schemes are gonna differ by industry some are gonna have more corruption some are likely gonna have more financial statement frauds talk to us a little bit about you know what are the most common schemes within the transportation industry and then if you can maybe walk us through some examples of each of those schemes that way we can learn a little bit more about what exactly you know corruption or maybe Billy what exactly does that mean and some examples that you've encountered number one on that list is billing now some people hear that and it can mean different things at different industries but for the transportation industry it for purposes of the AC Fe report that we've been discussing billing is what we're talking about there is purchasing accounts payable type frauds the most common thing that we've seen there are fictitious vendors or somebody perhaps maybe adding on to their house and they find a way to pay for those home improvements through their company so really again anything through the purchasing and accounts payable process so we dealt with one recently where we had a a fake vendor that was created by an executive and it was a losses were close to $800,000 over about seven years just as a just as an example and you know that's the transportation industry that's really all industries out there that are affected by the billing the non-cash side it's kind of interesting with the transportation industry it ranks pretty high but the best example there that I could give you is an equipment type theft right so we've had situations transportation otherwise where maybe tires for example or some other sorts of parts that may be valuable are taken from a shop taken from a warehouse and sold or somebody's got a business on the side so that's what you see there on the screen as number two is the non-cash type frauds and then number three is corruption we see this quite a bit which again the biggest things there are kickbacks so you could have somebody with a really tight relation ship with a vendor and in exchange for feeding that vendor business that it could be anybody on the inside of the company saying hey I want a piece of the action I want you know a thousand dollars off of every payment that I'm gonna make sure that you get right and that is just a pretty big problem all around and then number four is check tampering as the name indicates it's really just taking a check that you shouldn't and raising it for personal use so the big thing that we see there is somebody taking a company check honestly it's as simple as making that check out to themselves I mean it could be me sitting there I've got access to the company checkbook maybe I can sign checks so maybe I'm authorized maybe I'm not I put my name on it and I take it to the bank and cash it and again I know that sounds awfully simple but in the companies that have really bad internal controls that's what we see it's as simple as that you start looking at the bank statements and that's and that ends up being the problem okay and then number five on the list is payroll which what we've commonly seen are unauthorized bonuses and pay increases could be a payroll clerk it could be a payroll manager it could even get controller who has access and nobody else is looking at payroll right they'll go in and maybe give themselves a ten thousand dollar bonus maybe they then increase their base salary from 50,000 to 60,000 and nobody approves it right those are just those are the run-of-the-mill payroll schemes that we've seen so it sounds like really outside of the non-cash most of those are cash dispersements sites are really focusing in on where is the money going with that I'm guessing that the vendor file is probably a very important piece of the action so to speak when it comes to you know setting up a fake vendor or maybe a conflict of interest is that is that we've seen your experience the vendor file is really what it's focused on anymore it's one of the first things that we would look at especially when you get a company that calls and says we think we might have a problem or we want to see if we have a problem that we're just not sure so we would absolutely want to get our hands on that electronic file that vendor Master File the accounts payable detail file and really dig into that because the point you made is a really good one and you'll notice that cash receipts is not on there for the transportation industry what doesn't mean you shouldn't be concerned with it it's just that most trucking companies transportation companies are not going to be collecting a lot of cash you're gonna see more those types of frauds in restaurants and bars and that kind of thing that collective a lot of cash but yes for the transportation industry when and we get those sorts of calls where they say we want you to come and take a look the vendor file and the AP detail is critical right you mentioned that ap detail and it's interesting to check tam prints on this list I remember a case that I worked on just last year where an individual that's essentially what they did and in less than 300 checks in three years they get over a million dollars simply by making checks out to themself which seems so simple on the surface but we see it happen a lot and I bet you do it as well yeah and and that's I want to kind of build on what you said there when you said it's very simple a lot of these are very simple these are people who have taken advantage of their position and the fact that their company maybe doesn't have good internal controls or good monitoring and so once somebody like us comes in wheat are taking a look at what's gone on it is it's amazingly simple and it all comes back to nobody's really monitoring nobody's doing some searching some basic checks not even really sophisticated stuff right to see what's going on and is you know if the transactions that are occurring are they are they okay and we talked about the transactions in the monitoring and some of that basic stuff now it seems you know maybe 10 15 years ago the vendor file will be very difficult to review it was probably kept in a ledger a paper ledger where you know maybe you have it printed on green bar paper and you have the terminal that's two colors and you have to go through everything now though it seems like the prevalence of data is so much that all of this is something where you could have some level of monitoring actually focused on the data itself so in the last you know one to three to five to ten years how have you seen the role of data change in investigations and prevention work give a clue as to how old I am here and say that I've been doing some sort of audit or forensics work now for almost 20 years and full-time forensics were renowned for I guess about ten and to see the changes that have occurred over that time is just is amazing and it's just it still continues to change even ten years ago when I got into forensics work we were still spending a lot of time when somebody called and said hey you could come and help us we were going through bankers boxes what I'll do what I'll call wishful digging all right I think we've got a problem we need you to come take a look and so we're going through boxes we're looking for bank statements we're looking for invoices maybe not even one percent sure exactly what we're looking for we may be picking samples and really just in the last five years or so the data is gotten so good at many companies and often fairly easily accessible if you talk to the right people that it now plays the central role in many of the fraud investigations that we do and a lot of times this means not even having to go on-site people become concerned about causing a stir and they can send us a file with a million lines of data that we now have techniques that we can use to try to search for certain transactions and attributes and that kind of thing so I guess in summary yes we've gone from a more of a paper based sampling base you know wishful digging type situation to focused targeted data searches using electronic tools that that weren't available to us even ten years ago seven hours ago that are now there so it's it's really a great thing good as far as that data is concerned you know we talked about the vendor file on the transaction file and most companies are getting better you know getting that to you if you talk to the right person who do you typically have to talk to is that somebody in accounting or does a lot a lot of times do you have to go directly to IT it's a little bit of both and sometimes when we make that initial request you know after somebody says no we really need you to come help us we need you to come look and see what's going on with our accounts payable because we think we've got some problems purchasing we think we've got a billing scheme going on or again maybe it's just a check to see if they think you know if to see if anything may be happening that shouldn't and so when we make that first request a lot of times it may be to accounting and they'll forward it to IT and the initial reaction sometimes is oh we've never had to provide this before and we can get some pushback but I would say in 85% of the cases if we talk to the right people often we do end up talking to that IT director and the smaller company is the accountant the CFO maybe the IT director as well playing dual roles so it just it just kind of depends but normally it does land on the desk of like an IT Director or analyst who is able to get to that raw data that we want I mean that's what we don't want the PDF files it's the raw data that has basically everything and companies sometimes don't even know the stuff that's being stored there it's being captured that we can then take and mine and look for those transactions that may not be right okay and along the lines of that stuff that they probably don't know that they're capturing you know we've seen a huge increase in the number of digital devices that individuals have you know everybody probably now has either a blackberry or that may even be a bit dated I guess everybody has an iPhone or a Galaxy S whatever number it is you we see everybody carrying around devices sometimes three or four or five devices and the transportation industry communication devices are everywhere whether it's on the person or in the vehicle how have you seen these digital devices play a role in investigations I would imagine that in some capacity you can get email data text messages data and use that in your investigations in some way yeah absolutely and it's become a just as we talked about the accounts payable and vendor Master File information that type of data that you're talking about which I'll call globally digital forensics is becoming increasingly helpful I read an article recently about the death of the phone call everybody's communicating now by text message and email mm-hmm nobody's leaving voicemails hardly anymore nobody's picking up the phone everybody is communicating electronically which for us when we're trying to figure out what's going on behind the scenes is incredibly helpful we've been able to recover deleted text messages photos address books off of iPhones things that people thought were long gone on the tablets well I guess I'll go more talk about laptops any sort of basically digital device getting emails off of those devices has been critical and helpful I think we maybe even have an example that we can show the alien definitely this was one that that I'd like to share with with some audiences and when I'm talking about fraud schemes it just illustrates how maybe a seemingly innocent email ends up being a critical part of a kind of uncovering the total fraud but it says hi so and so we've not seen any progress on the bank statement reconciling items discussed at our meetings several months ago we understood that they would review the items and work to clean up the reconciliation we need some resolution the account the audit begins in two in two weeks and I would like to have these items cleaned up please let me know if this cannot be accomplished and what this ended up being was just the tip of the iceberg it was actually over a 1 million dollar fraud Wow and so part of my message here is not only to tell people hey if you've got a small discrepancy in your records you need to maybe keep digging if you just can't ever get it fixed because that you could just be seeing the tip of the iceberg and then this instance it was a fraud where this particular person couldn't cover up the complete scheme and so there was always these reconciling items on the banks on the bank account but this was one of many many emails that we found and and we could spend a lot of time talking about the text messages and everything else that can really help put a case together tells us who was who was talking about what and when oh really critical stuff and I would imagine the fact that so many people you know they sync their iPhone to their laptop or they just unknowingly plug it into the work computer day in and day out to listen to music or charge it or whatever the case may be that that's information possibly being captured that it's now on a company device and could be a fair game in these investigations yeah and most of devices that we're talking about here are typically company owned I mean we're not law enforcement we're not going out and issuing subpoenas are going to homes and collecting iPhones a lot of times it's company property and it is fair game and when you have those we typically try to like to get our hands on them and try to see what's on those devices it's because again it is just tremendously helpful in understanding kind of what's going on behind the scenes and who's talking to who okay so it sounds like it's definitely much more than just our financial data that we've traditionally thought from an analytic standpoint it's not just you know the debits and the credits from the general ledger but it's getting all of the data that's available to us and you know keeping in mind that there's a lot of data out there that we probably have never thought of and using an investigation that could be really useful to us and detecting frauds that have happened in the past yeah so I mean we've talked about the accounts payable and the vendor data you know payroll data employee data and then the text messages and emails we've even gotten into some I want to call them non-financial measures but other reports unique to the industry for instance about a year or two ago we did a project for a trucking company that had some concerns about fuel usage on trucks and cards and so we actually got a file that was I don't know 60,000 or more lines long and we're able to go in and do some analytics to look and see is there anything about certain car fuel cards that just didn't quite make sense compared to others so yeah I mean if the data is there we can typically use it right and to try to figure out is there something amiss it's definitely worth considering yeah if it's there use it yeah right and it's one piece you know each of these plays a part that we've talked about it's rarely just one thing it's that bank account it's the fuel reports it's the Accounts Payable information it's those text messages from a phone it's really what I call putting a puzzle together okay so when you have all those pieces which if we're lucky enough we do have all those pieces that we can put together normally we're not quite that lucky it's just right little selected pieces but when we do have all of them that we can it helps create a picture of what happened and then we can go and help whoever maybe figure you know explain what happened okay it could be having to tell them that they've lost you know half a million dollars to a fraud scheme and here's how but I mean we don't like to do that but sometimes that's part of it yeah I would imagine that's probably one of the more difficult parts is having to explain by an owner that they've they've been hadn't really essentially yes yeah well talking about you know this picture and then explaining to them you know what has happened what are some of the major weaknesses that you see that allow these frauds to occur probably the number one thing that I personally seen is just poor segregation of duties what I'll call employee a doing everything from A to Z that's really one of the bigger problems and I know the ACO fee I think we've got a slide that we can share to show some of what the EC if he has found any way and it pretty much lines up with with kind of what I'm talking about here and the number one thing on that list there is a lack of internal controls and that's what I'm talking about is where the same person can write checks sign checks approve new vendors approved payroll just again just doing everything from A to Z and that's generally too much the second thing is lack of management review I know owners get busy they're out making sales they're traveling they're trusting their employees to do the right thing you know and I know the world revolves on trust but what what I found there is you get really almost a complete lack of management review they're not looking at the critical things they need to look at and things that really probably aren't even large investments in time but they're just not looking at the reports they should they're not looking at bank activity um that kind of thing so like you could see that generally lack of management review and the lack of internal controls are really a couple the biggest weaknesses that are out there okay you said that a lot of times Trust there's a lot of trust that goes on in the organization I haven't heard you lose trust as an internal control no per se I mean I would think that we all want to hire employees that we trust but there's an element there of responsibility over those that we trust because that's a lot of times what we see I mean it's the most trusted loyal employee that's been there forever right they can get by with all of this is that typically what you see from a fraudster the trust factor is one of the biggest pieces to all of this we rarely see anybody who is been there a couple years sometimes we do or a year to start a fraud scheme it's normally somebody who's been there at least five maybe ten years sometimes 20 years they may have never done anything wrong in their life and maybe have never stolen a dime from any any place I've ever worked and they maybe they get in a bind or they see something they want or maybe they see others that are making more money and they feel like they deserve and a lot of rationalizations can go on there but for some people it's not that far of a leap to decide you know I'm going to want to take this dollar and I'll pay it back that's the most common thing is it's a loan I want to pay it back later but what happens is nobody hardly ever pays any of that back but that's part of the rationalization of it they just kind of continue to take and keep thinking eventually one day eventually eventually get that day that day almost never comes okay alright Jeff we talked a lot about you know some of the general trends in fraud and we've talked about how data has changed you know its role in fraud investigations we've talked specifically about frauds and transportation some examples of cases that you've worked on and how you performed various investigative techniques using data with that it seems to me that a lot of what is used as an investigative technique on the flip side can then be used as a preventative technique going forward and kind of using that okay here's what we've had happen to us in the past now let's make sure that going forward it doesn't anymore we're gonna look for those specific patterns and with that it seems to me that data as it's growing in its role should also be pretty prevalent when it comes to fraud preventative controls going forward now saying that using data is gonna prevent all fraud there's really nothing out there that will but how have you seen organizations start to use data as a preventative control are they using data out there as a preventative control and is it effective if they do a few are you not not not that many still and what I've seen is companies that have been victimized by fraud then tend to make it a preventative control know what I'll call a detective control because that's really what we're almost talking about is a control that would help pick up things that maybe before they really become a big problem so once somebody has a major problem in it aside now we need to do something to try to fix it and you do get companies out there that are being proactive and saying we need to try to figure out if we've got problems or maybe seeing other have heard from friends you know they're also in business about a fraud they've has now they decide they need to try to figure out a program that will help them monitor for again fake vendors payments that weren't authorized and I think we have a graphic that we could show the audience that this is again from the 2014 ACF e survey and if I'm if memory serves I think this is the first time the data analytic shows up as on this at least in terms of how effective it is right I think as a control in terms of reducing losses anyway shows there on the right-hand side that having a proactive data monitoring or analytics program is associated with a 60% reduction in fraud losses Wow so what that's telling us is that if you have a good data monitoring analytics program you can pick up on those things before they become the 10 and 20-year frauds that I mentioned before right you may cut a loss from it was gonna be a hundred thousand dollars that may become a fifty thousand dollar loss okay there's though it just unfortunately seems like companies are still reluctant to jump into this they don't understand the data analytics they got too many other things going on but once you get a program rolling and in place it's can almost run in the background it is so much easier than somebody again sending in and maybe an audit team once a year to dig through things it can be done without auditors on-site a lot of times and it's just and I think that's probably one of the main reason that you're seeing it's one of the top new methods of detecting fraud matram is effective controls is for that reason it's just become with all the data out there it's just it really works well right and it seems like if you do have one of these you know techniques in place that it's more of a preventative or even a detective control then when you are having to follow up on transactions and maybe pull a sample of 30 or 50 or whatever approach you may want to take at least you're pulling that sample from a set of exceptions you know it's more of a focused effort here when you're asking the questions you're asking better questions and that probably goes a little bit toward that perception of detection where you know people are employees or thinking wait a second they're not just asking Oh what about this check or this expense reimbursement and there wasn't really no premise behind it right now it's more they're asking about an expense reimbursement or you know a fake vendor or any type of vendor and they're asking very specific questions that probably raise the eyebrow of those that maybe see that opportunity and make them step back and say is this really an opportunity sure yeah there's a theory out there and you mentioned it call the perception of detection meaning if an employee knows they're being watched and may may even even if they think they are a simple example fake video cameras or security cameras around a room if nobody knows that even if they're fake people may be less likely to do something wrong and so with the data analytics yeah I mean you can you can call a say 1 million line file so I want to get 1 million disbursements a year right you might be able to whittle that down to 200 transactions that you're really interested in based upon certain attributes its trends that don't make sense it's some vendor names that you don't recognize its addresses that just don't quite line up with what your expectations are and so when you get that list down to say 100 you could go around you can research those vendors you may be asked some questions of employees and it's just it's just a really nice complement if not a key part to an overall fraud risk management program as far as using analytics as Detective controls and we're going to kind of go back to the investigation side I think you brought a couple examples of how you've actually used analytics in fraud investigations could you walk us through a couple of those sure yeah I think it's always helpful to see some actual examples because I know we've been talking here for a while about different types of files and there's probably people out there scratching they're insane what are they talking about right let's kind of through a real example here so there on the left hand side the yellow section that's an excerpt from a again from an actual file that we had that we were working with trying to look for vendor employee matches so the yellow information there came from the vendor file an Accounts Payable file and then on the right hand side there that's employee information so you can see the yellow highlights there there's a ridge crest investments and in all goods lawn care both of those matched two different employees so the concern there would be is this a fake vendor if it's not a fake vendor why are these employees allowed to do business there basically they're collecting run a check and then they're also they've got a side business are they really performing these services for the company right now most of the time what we find is that these Inns these end up being not necessarily fake vendors but somebody doing business as an employee so they're collecting their paycheck they've also maybe have somebody maybe maybe it's their son maybe it's themselves let mowing lawns for the company on the weekend collecting right an extra check now is there anything inherently wrong with that no not always but when you maybe ask the executives and management of a company do you know that your employees are doing these things and they say no we had no idea so it just it really helps just even from a from a company policy standpoint understanding what those relationships are with your between vendors and employees so it sounds like there are some definite benefits outside of just fraud detection I mean compliance policy enforcement conflict of interest I mean learning more about what's truly going on in your company versus maybe what the policy says is going on or what people think is going on the data is going to tell you what's actually going on a quick question to follow up on this example here you know that all goods long care it's 13,000 dollars roughly in payments over a bit of a time frame that seems right off the bat I don't know the company obviously but to a lot of people out there 13,000 dollars over a year maybe 18 months is probably gonna seem fairly small you know is that really something to follow I talked a little bit about how you know this small amount while it may not be a million dollar fraud today could grow into one later that you know I'm guessing most frauds don't start with somebody taking a million dollars the first try right yeah they start small so you know in this particular case you don't know until you start digging you have the start of something that's going to grow into a bigger problem and and so if we're lucky we catch something like this on the front end where we say hey you've got a fake vendor here and it looks like it's been going on for the last year but maybe if we look at that if the the trend it was five hundred dollars the first month it was a thousand dollars the second month and then it became three thousand dollars and then it became five thousand dollars well project out ten years fifteen years if you do if nobody does this test to figure out is there a vendor out there that we're not aware of and an insider the table to influence that and to send those payments through the sister the Accounts Payable system and it's a pretty scary thing so yeah you can't just immediately look at the dollars oh it's not that big of a deal you really need to understand the relationships and is it okay was it approved was it run up the chain so yeah it's it's really really really helpful okay and I would think that it's a whole lot easier to say that well you lost thirteen thousand then you lost one point three million right much easier conversation to have yeah but you have to have it much easier yeah and you brought another example to correct yeah I believe I believe we could put that up and I know earlier you were talking about what are the changes in data analytics and and just the kind of the neat stuff that could happen now in terms of discovering relationships between vendors and employees this is one of those things that even when I started I got out of college now I get almost twenty years ago but I never dreamed was possible thanks to the invention of Google and satellite mapping so what we have here is when we took a particular clients file we got their employee master file and we got their vendor master file we can actually assign GP coordinates to those addresses can plot them on a map hmm and you say okay well that sounds all really neat techy and all of that you know but why is that so important when you it but it's interesting whenever you start plotting these where you see relationships maybe that don't make a whole lot of sense and so here's one now granted we've dummied up the names to keep it I guess anonymous provided to go to protect so we've got a Vinnie salvage are there apparently has a pool which is interesting for a salvage yard and then you've got the AP manager living there kind of caddy corner from then a salvage yard now there's almost no way that you could ever discover that if you looked at all invoices if you picked a sample I mean even even if you had picked a sample and you'd selected Vinnie salvage yard what could you possibly tell from the invoice I mean you would have no idea that the address is very close to the Accounts Payable manager now right here again I want to offer some words of caution because it would be easy for somebody to look at that go oh there's all it there's definitely a problem the idea is to take this information and use it to figure out is there something going on you can't automatically conclude based upon this that there's something weird or something improper or there's a relationship but certainly you know next lines of inquiry would be does anybody work with Vinnie salvage yard who's the principal's who are the owners who are the employees who we deal with are they really providing services so this is just it's it's a really it's a it's a really really helpful tool again thanks to the nice technology that's available today right through the internet so it sounds like something like this and a lot of the tests we've talked about so far it's very helpful to use data analytics to tell you there may be an issue in this area but it doesn't sound like data analytics really tells you this is fraud just write the report now there's an additional step beyond just looking at the data right we may have a file again this example really has a million lines in transaction detail and I like to ask question which would you rather do you want to come in and randomly pick a hundred of those or maybe even 500 and camp out here and look at those invoices and hope we find something and honestly even if we pick those invoices there could be something wrong with that relationship that we wouldn't pick up on so you know but the flipside is let's take that file and let's call it down to one hundred two hundred even three hundred different transactions right that are focused again based upon the trends of that vendor how much has been spent over the last three years the addresses the tax ID numbers let's let's look for those attributes that we know from other frauds might be apparent so when we get that list then we can start digging into those and there's a good chance all of those may be okay right but do you want to take the risk and you know just hope that there's no problems there so it is such a better tool again that when I started out in doing some audit type work which is you know let's pick a sample and hope that we find something right same point for fraud is sample our drawback t'v right okay for fraud if you're trying to find those vendors those employees that may be fake we haven't talked about that but ghost employees even in a vendor in the employee Master File it's it's really one of the best ways that it exists now and the technology is there to like to make it happen okay well we've talked about a lot of great techniques you know had the examples we've talked about the common schemes which really focused on the cash going out the door and the disbursements on the you know billing corruption conflicts of interest kickbacks etc so what within the transportation industry what are some of the best files and maybe this is true across in you know any type of industry but what are the best types of files when it comes to a fraud investigation to kind of focus in on these that we've talked about billing corruption you know what are the files that have you know a company is looking to start analytics what do they need to get the key files to start would be a vendor Master File an Accounts Payable detail file an employee master file I know sometimes there's concerns around Social Security numbers and you know confidential employee data but you know typically care can be taken to make sure none of that is compromised a payroll detail as well now definitely if there's concerns I mentioned earlier we had a project where we're looking at fuel card restitch kind of the sky's the limit and you know whatever data is tracked out there if there's some concern about certain drivers let's take the transportation industry or loads or routes or whatever it may be if the data is tracked which from any trucking companies it is right we could take that too but I mean the core files are the four ones that I'm in surely which has been a master file Accounts Payable detail than employee master and the payroll detail files okay and you touched on fuel cards I mean in my experience at least on corporate credit cards purchasing cards and fuel cards a lot of times those are some of the best data files because they're coming from a bank you know you know relying upon somebody to manually enter that data but that system generated at the financial institution level and those are a lot of times very data rich would you agree if you can get information from a third party and we've had those where it's purchasing card data it's bank data of various types when they can give us that information directly and then we've got even a better source of data that we could mine and analyze to say again are there trends here by user by card user that just don't make sense that's that's those are just tremendously helpful okay well as far as outside the data realm I mean obviously this webinar is focused on fraud prevention and detection the case for analytics so we want to focus on data and I think we've done a lot of that but at the same time it seems like having an expert such as yourself on investigations and prevention work it would wrong not to ask you about other anti-fraud controls that organizations can implement in their organizations so if there were a few others out there control wise outside of you know proactive data monitoring what other controls would you say that organizations should definitely consider probably one of the primary ones that I always like to recommend for and this would apply mainly to some more than medium-sized companies and larger companies is a hotline tips are actually the number one way that most frauds are detected so your employees and may even vendors and others are your best eyes and ears there there may be somebody working in a cubicle and they're there person sitting across the aisle from them maybe is coming in late and they're acting kind of strange and there's you know some transactions going on that they're not quite sure about maybe that person knows something and they don't know how to report it so you know we'll pick up a phone or have a different confidential reporting mechanisms as important so hotlines again we talked earlier about segregation of duties and not allowing one person to do just everything making sure that you're dividing up the duties don't let the same person write checks and sign checks have somebody else reconcile the bank account that's critically important and today we've talked a little bit kind of about some fraud type education I guess topics and so educating people your employees on the warning signs of fraud how they should you know feel like they could come to their supervisors or they've got a hotline they know that they can use so some fraud education so people know to recognize these things that we've been talking about there's other things but those are the probably the three critical pieces that I would mention it okay as far as warning signs if there were you know a couple warning signs out there that are the ones that you encounter the most where you've talked to a company you've done the investigation and then you're you know telling them I hate to tell you that you know it was a five hundred thousand dollar fraud inevitably it seems like they're likely to come back and say well you know thinking back you know we this employee do a B C or D probably should have thought about it but never did what are some of those warning signs that they should be just on the lookout for yeah the number one thing that I've seen and I think this bears it out in the data and the surveys that we've been talking about from ECFE lifestyle changes is actually the one of the number one red flags if not the number one red flag okay so this is somebody that's making thirty thousand dollars a year that's they're going to Vegas six times a year they took a you know around the world cruise and they drive a Corvette and you know the story is that they've inherited money where their spouse makes a lot of money those things may be true but it's probably worth looking at that's so lifestyle is the number one thing when they when people look back and they say this was a warning sign I missed that's that's that's a big one people who still typically don't like to take vacations either so they may be the hardest-working employees in the office because once you start a scheme it it's hard to keep that covered up when you're not there so they always find a way to come back to the office and we've seen this before people were supposed to be out of the office on vacation but they found a way to bring masais cookies and cakes back to the office every day right and check their email to make sure that nothing weird was that or that nobody had discovered what they were doing so lack of vacations people who insist on working weird hours so we had a lady that was counting cash and she liked to come back and count they allowed her to count cash from 7:00 to 10:00 p.m. with her with with a relative at night and it was just the two of them and so it doesn't sound like they might be a problem yeah so again it's unfortunate because you know fraud education understanding these types of warning signs can can really help if sometimes it's just too late right okay well Jeff thank you so much for your time today we really appreciate it I think like I said having an expert like you with your experience and the variety of investigations I know you've worked on is really valuable for the audience so thank you very much for your time oh great thank you thank you alright as we've moved on throughout our conversation here jeff has provided us a lot of great information in insights about investigations the transportation industry and just generally how data can be used for investigative and preventive purposes so now we're going to kind of wind things up going back to our application framework I'd encourage you to take what you've heard over the last 30 to 45 minutes and really apply it within this framework think back to phase 1 what exactly is the risk that you have in your organization after hearing some of the examples and the stories that Jeff has provided you probably have an idea as to something that may be happening in your organization so go ahead and let's plug that into phase 1 then let's start talking through phase 2 how are you going to design objectives they're going to help you address this risk those objectives maybe some of the tests that we've talked about some of the examples that we've provided but those objectives are going to be very specific to your organization once you have those you're going to need data and Jeff has provided us a lot of ideas as to the data we may need whether it's the vendor Master File Accounts Payable detail employee master payroll detail or our fuel card transactions we're going to have data and we're gonna have a lot of it and remember what Jeff said that it may not be the first person you talk to when you try to get the data that ultimately helps you get it you may have to go to multiple sources and if you can go to a third party by all means feel free to do so once you've done that you need to start developing your procedures go ahead and break those down to the basic level and step through them chronologically one procedure after another and once you've done that then you can assemble them into a much lower much larger more complex set of procedures after that you're on to analyzing the results again this is something that we've all done we may not have done it on a hundred percent of the data set but you've analyzed results in a number of capacities already in your career now we just need to apply that to the data side so we want to start analyzing our results determining if we've met our objectives if we've addressed our risks and how exactly we're going to use those results going forward and using those results going forward as phase six we need to manage those how are we going to get those back out to the business so we can make better decisions run our businesses better and better manage our risk I'll go ahead and leave you with a closing thought and that's when it comes to analytics you have to be willing to try a lot of organizations don't do analytics because they're just simply afraid to get started and you have to be willing to give it a shot I know in my career I've created numerous analytics that weren't successful but the key is I learned from those failures and was able to build a better analytic the next time so that would be my recommendation to you don't be afraid to give analytics a shot failure can definitely be a springboard for success even in the world of developing analytics the key is to learn from those failures develop a better analytic the next time and really have an impact in your organization by leveraging data to better manage your fraud risk thank you so much you

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *