Select Page
Boxes with one red
ON-DEMAND WEBINAR

How municipalities use AI to reduce the risk of procurement fraud

With new funding programs, infrastructure activities, and uncertainty in a COVID-19 environment, staff within procurement and project management divisions are faced with unprecedented incentives to influence activities or leverage systems to make financial gains or cause loss. This challenges municipal governments to find new ways to identify and reduce the risk of procurement fraud.

How can financial leaders use AI to understand their data and augment fraud detection activities?

Join Kenneth Pun, CPA, CGMA, Managing Partner, Assurance Partner, The Pun Group, LLP, and Michael Bottala, CPA, MindBridge, for an expert walkthrough of how AI reduces fraud risk and maintains taxpayer confidence. In this webinar, you will learn:

  • Types of procurement fraud and the differences between fraud and mistakes
  • AI and machine learning techniques used to detect fraud
  • How Transaction Analysis Services reduce fraud risk

This webinar includes demonstrations and a Q&A session to answer your questions.

You will walk away with a better understanding of procurement fraud and the value of Transaction Analysis Services in reducing fraud risk.

 

Who should attend
  • CPAs and CAs interested in procurement fraud
  • CFOs, controllers, and internal auditors interested in procurement fraud
  • MindBridge Ai Auditor users

 

%

of US companies experienced fraud in the past 24 months

%

of US companies do not have a formal bribery and corruption risk program

Transcript

Kenneth Pun 

Good day everybody. My name is Ken Pun. I'm the managing partner of the Pun group. And along with me today is Michael Bottala from MindBridge,  How can municipal finance leaders use AI to reduce procurement fraud? This is a co host between the Pun Group and MindBridge who is a certified public accounting firm specializing in the state local government practice and also MindBridge is a AI technology firm and designing the software to help auditors just like us do a deep dive on the analysis and helping the state local government here so what we are going through today is is going through the types of procurement fraud that we can identify in the state local government and also identify what's the difference between a fraud or a mistake and MindBridge will come into play Michael will talk about,  they In the last couple of years been building up these nice AI software tools and how these machine learning technique be able to use it in detecting fraud. And then also last but not least, is how we are able to help your organization to reduce the fraud risk. Fraud is actually defined by the Black's Law Dictionary as a knowing misrepresentation of the truth or concealment of a material facts to induce another to act to his or her treatment. So consequently, the fraud include any intention deliberate acts to deplete our another property or money by guilt, deceptions or other unfair means. So that mean like the difference between a fraud incident versus a mistake, basically is the intent whereas the intense are like, the mistake is more of an unintentional and the fraud is more of a intentional. The fraud triangle is really your compass for preventing and detecting fraud.

 

Kenneth Pun 

The three elements here including the opportunity, the rationalization, and the pressure, kind of go going through the, the each element here, opportunity really refers to circumstances that allows fraud to occur. It is the only components unit that the organization exercise, complete control, basically a couple of things that we can see, while people giving, like having the opportunity of fraud, basically, including weak internal control, poor tone at the top, and then the inadequate accounting policy here. So the other one is what we call pressure because in a government organization, like not normally you do have like in another word, there's some incentive but in government sector, we use pressure instead of incentives. In this fraud triangle because the state local government usually don't pay bonus, but a lot of time is on the on the personal side, the employee themselves will see that they, they experience a lot of pressure, like financial pressure for their own personal and last but not least is regarding the rationale. I'm saying that hey, this is where the individual will justify that for them to commit fraud. Are they even telling the the organization I've been putting all these hours in and they treat me really wrong and not paying me well, and why the upper management keep getting all the money and this is really the drive for people like committing fraud here ACFE every two years they do have a survey. And the last survey this is on the global study on occupational fraud and abuse throughout the whole entire world. One of the thing that they do pointed out is 5% of the revenue are basically to two fraudulent transactions each year. So keep that in mind like in an in a governmental entity that you really not focusing on on revenue, because majority of the fraud transactions do happening in the in the procurement side. So that's why when we when we first talked about what kind of webinar that we would like to focus on, is it for the government for the governmental arena here is really focusing on the on the expenditure side and and focusing on the procurement fraud. So think about if the agency has about like 100 million dollars of like it like on the on the appropriation, budget and 5% of that is equal to $5 million and $5 million of public funds, which is a big dollar amount here, the most common form of fraud in the government sector is the fraud in procurement.

 

Kenneth Pun 

Why we want to say that because government is not really a revenue generating and kind of make profit out of the out of the taxpayer, but the thing was government contract, there will be a lot of opportunity to play it with the government, including kickbacks, conflict of interest of fake companies and orders, and the and fraudulent invoicing and collusive bidding process. I can tell like, back couple years like backing upon like 2014 I can give you a perfect example here. City of Pasadena there is a former city employee and a contractor basically, like doing colluding together and trying to steal about like $3.5 million from the city, one of the Public Works employee actually have like is even charging of their utility line on the ground and basically having a whole lot of work to be done in, in in the area. So the person the employee actually creates a lot of like fake invoices for the program and submitted to the to the purchase like to the AP department, and at the end of the day, he stole about like $3.5 million and also this guy actually follow through some of the money basically giving kickbacks to the companies that he worked work with called Colin's Electric. So basically because you are doing utility underground electric works That needs to be done so is perfect example of working with a outside vendor and providing some kickbacks and and funnel that that that money to the to the third party. So having said that like kickbacks conflict of interest corrupted influences and collusion in and manipulation by biters because basically when you actually get that types of work you got to have a your influences influences in selecting the the bitter so in this situation they like even though I don't have a whole lot of detail but from the from newspaper, but this is definitely a perfect case to basically talking about the different fraud scheme that people will use surrounding the procurement areas. So a couple of the other things that that when we talk about the corruption Within the procurement fraud, some the basically they can work with the vendor and basically leak out confidential information. So, what does that mean is they already received the the bids and basically kind of leaking the the bid prize the the and also the qualifications to another vendor so that they can submit the bid appropriately. Sometimes even the when they design the RFP, the request for proposal, they will actually tailor made it for that particular vendor. So basically all the SPAC and all that all the the scope of work, it's it's really focusing in improperly favor that one particular vendor. And then also when the bidder actually when the vendor actually submitting that bid and they will have a light go on. Through a panel and and scoring, and basically they will have all kinds of bias and then basically put the vendor at a very, very high high score to give an advantage to that one particular vendor. And for some incidents they they'd likely to award the the contract as a sole source and kind of like telling telling everybody that because like even even looking at the specific others the SPAC they may not be able to arm basically favoring that one one vendor and then at the end of the day that one vendor came in as a sole source contracts. So this is trying to avoid any competition here. So overall, when we actually dealing with all these type of procurement fraud, we do see couple things one is lacking management overview. 

 

Kenneth Pun 

Most of the government agency do not have any internal audit function within the organization. So, there is always lacking the internal audit function, sometimes smaller agency, you do have a resource limitation. So you will not be able to cross chain and other personnel and then share the duties. And then last but not least here is very important, they are lacking the independent data analytics or ongoing monitoring process. So, that that is coupled the warning sign of a potential fraudster like especially for employees to committed fraud, you'll be able to tell like 50% of the fraudster display that they are living beyond their means and 36% of the fraudster basically they have some sort of financial difficulties are 22% of the of the worker basically, they have a tendency of controlling and unwilling to share the duties. 24% also show unusually close association with vendor and customer 18% across their work era, irritable, defensive and suspicious and last 10% is refused to take vacation. I'm not quite sure some of some of you guys actually look at knowing that there is a case study and even on on the on the Netflix app episode. I'm looking at the city of Illinois, the Illinois no city of Dixon in Illinois. And there is a one big case talking about the controller stealing about $54 million out of a city over the last 20 years to build up that her horse ranch. So this is is a perfect case to think about like who this, who the fraudster is. And and and some of the warning signs or shown in that one particular case, how are transactions analysis service reducing fraud risk. So, one of the things that we need to understand is the need to keep up with analyzing your data because I know that there are a lot of agency they analyzed it the data, but without really looking into the data and proactively responding to the risk that they identify in this analysis. So what we as the parent group, we designed the the transaction analysis service for our client and teaming up with MindBridge here to perform couple things. We basically will we'll go through the data monitoring process and identify and extract the relevant financials, financial data from your general ledger. Also in the meantime, this this year on MindBridge also have the AP module which further further expanding the scope within the the purchasing module. So we extract, transform, and load all the information to the AI software, the MindBridge AI software and perform a local review. So in the meantime, like when, when, like in the next slides, Michael is going to go through and show you a platform and show you some demo and see how this software will be able to identify risks. Once the analysis is coming out. We as the  auditor coming in and look at that analysis and performing additional analytics. And last but not least, is to interpret the result and providing the CFO, the finance team and even the commissioning body On what what is going on within that organization and then providing our suggestion to the, to the organization to reducing the risk. So, Michael, I'm going to turn the floor to you to talk a little bit about the software here.

 

Michael Bottala 

Okay, great. Thanks, Ken. Okay, so Ken, Ken did a great job of introducing the product. So we our focus is on the general ledger, AR and AP modules as it relates to the government. Specifically, the general ledger and AP are very applicable to a lot of the things that Ken mentioned earlier about asset misappropriation, helping management oversight, and data analytics as a continuing continuous monitoring service. So we'll talk a little bit about the MindBridge approach and how we help analyze information to surface those anomalies in those unusual situations. In order for professionals like Ken to to you use their expertise to make judgments about this information and how it's impacting your organization. So as you can see here, this is the MindBridge platform.  This file has been uploaded, it's a general ledger analysis, all the all the data has been analyzed. And this is the what we produce as the risk overview dashboard. So as Ken mentioned, every single transaction in entry is ingested into the system and bumped up against our control points. So these control points are a mixture of business rules, statistical models in machine learning. If we go ahead and dive into a high risk transaction, we can get a better understanding of the specific control points that MindBridge offers. So what I've done is I've gone into a high risk transaction, I get visibility into that exact transaction, and if I select any of those line items within the transaction, I'm now able to see which control points are triggering for that specific line item. At MindBridge we combine business rules. So we look for things such as cash expenditures, duplicates, reversals, suspicious keywords, then we layer on statistical models such as two digit benford, which basically indicates that you would expect to see a certain number show up a certain number of times. If it's showing up more or less than anticipated, it could indicate human manipulation, and fraudulent transactions. And then we also layer on statistical models such as unusual amounts. So we're looking at each unique account within your ledger and understanding what's normal on a statistical level within that account, where MindBridge really separates ourselves is the machine learning components and the ability to uncover the unknowns within any data set. So the way we're able to better understand those unknowns is through control points such as expense Flurry.

 

Michael Bottala 

 So what we're doing here we're understanding each different Fund and the volume of transactions to that particular fund accounts at any point in the year. So for example, if there's typically, you know, five invoices to your repairs and maintenance on a monthly basis, and then all of a sudden there's 15 or 20 transactions going to those particular accounts. Now we're seeing an expense Flurry, which could indicate that there's stuffing of expenses and improper approvals happening because of the increased volume in that particular area. Similar to expense flurry and understanding the activity, we have what we call split expense, to split expenses actually looking at if, if an accounting, an accountant within the organization is recording an entry to multiple different funds. So for example, sticking with that theme of repairs and maintenance, if now we're allocating it to one fund one, fund two and fund three, we want to make sure that our allocation is done correctly and the proper approvals are taking place for that specific split expense. Another key component to MindBridge is understanding the relationships between the accounts that are being recorded or the transactions that are being recorded and the accounts that are being impacted by these transactions. So the way we're incorporating that is through control points, which is rare flows and outlier anomaly. So rare flows essentially looks at the frequency in which particular transactions are occurring within the ledger. So we're able to understand what's a normal type of transaction and which accounts should be impacted by that transaction. For example, paying repairs and maintenance expense through your payables understanding that's the normal way that that particular account has transactions recorded to it. And if now we're seeing debits to repairs and maintenance and credit to cash. Now that's outside of the norm, and we want to make sure that there's proper approvals taking place If it's coming straight from the bank account. Similar to that approach, we also have outlier anomaly which now understands the normal amounts that are being recorded to each account. So to summarize all these different control points that we offer, it's essentially looking at the frequency, the relationships between the accounts and the amounts being recorded to give a better perspective on what's normal for this particular ledger and identifying any unusual situations or anomalous activity within the transactions being entered. So now that we kind of have an understanding of how these control points are being triggered, what the system is looking for, let's talk about how it's presented within a transaction analysis service that Ken be offering.

 

Michael Bottala 

So all this analysis has been been done and now it's up to Ken to use his professional judgment to understand how these can impact the financials. And if these mistakes can material miss could, lead to a material misstatement of your financials or if there's fraudulent behavior going on. So within the high risk group within this risk overview page, we're able to get a strong understanding of the type of transactions that are coming through because we can see all the control points being triggered, we can see a stratification of these transactions over time on a monthly, weekly or daily basis. So we're able to see if there's any unusual spikes in the activity or, or maybe there's a week with lower volume of transactions coming through, we're able to understand the norm in terms of the volume for that particular fund or organization. And then we're able to break down the accounts and the risk by user as well. You know, we mentioned earlier about the opportunity in that fraud triangle, and part of the opportunity is the controls within the organization. So if we can get more visibility into who's entering transactions And making sure that they're entering transactions within the scope of their work, we're able to, you know, reduce that opportunity that somebody within the organization has the opportunity to commit fraud or, or, or, you know, misappropriation of the assets within the organization. Part of the screen is what we call a categorical risk filter. So you see that this is a nice 10,000 foot view of everything that's going on. But what we can do now is we can break down specific areas of the organization. To get a better understanding of where the risk may lie, we can select a specific fund, and everything on this page is going to update related to that fund. We can also select specific accounts. So if we go in and we go ahead and select our accounts receivable, we would expect we can see the activity within that account the type of transactions that are coming through, as well as which users are entering During transactions in that area. So right now there's quite a few users that we're seeing within the AR, account. And maybe these are all different AR clerks that work on a part time basis. But it seems like there's a high number of individuals here. So maybe we want to better understand the controls that are in place in this organization, and start limiting the approvals of what's happening within that specific account to prevent that opportunity from occurring.

 

Michael Bottala 

The other key piece is understanding the type of transactions that are coming through. So we can also filter off of a specific type of transaction. Maybe it's related to manual entries. So we can select our manual entry code, we can see that you know, it looks like there's one entry reoccurring on a monthly basis. And we can also see what accounts that that particular manual entry is affecting, you know, within a quick eyesight so we can can now use this judgment to say are these the appropriate accounts that should be impacted by that type of transaction. As we move forward here, we can move into our trending screen which which offers the ability to break down these accounts over the course of the year. Because of that, we have that underlying General Ledger information. So we can see the the ending balance of our assets or the activity over the course of the year. What we can also do is we can compare and do and look at different relationships that may occur within the ledger. For example, if I wanted to see how my assets compared to my expense activity, now we're able to see that on the single graph and get an understanding if there's any unusual spikes, or unusual trends that don't follow the typical pattern for this or for organization. What we can also do within this trending screen is break down the different funds. So we can select funds that are maybe related to one another. We can select a specific account, and now such as expenses, and we can understand what the activity looks like within these funds. So we can see there's a big spike right here in terms of activity. You know, maybe we want to dig into that specific time period and get a better understanding of why it's, it's different in this timeframe versus this timeframe. As we scroll down, we also have the ability to incorporate key ratios. So with any organization ratios are a crucial part of understanding the history of the organization and whether or not we're operating within kind of the normal functionality. At MindBridge, we offer a proprietary algorithm called saaremaa. So saaremaa takes a moving average of five years of data to develop an expected range for your organization. From that expected range, we're able to compare that to actual results and highlight any outliers and then we're also able to forecast where we anticipated In this current period to be based off of the history of your organization, so it allows us to pinpoint those periods of time where the operations differed from the past. And now we can dig into that specific area and see, you know, why is this different, did did certain approvals take place, we did not anticipate that we go beyond the budget in certain areas, you know, now we're able to have visibility into that. 

 

Michael Bottala 

The third key piece of our program is around the data table. So once we identified the areas that we want to look into, we can now break down the area in any which way. So we can break down any individual account, we can build in specific types of relationships. So if we wanted to look at things that are increasing in expense account and decreasing our cash, we can go ahead and build that particular relationship in to have more accountability for the cash that's leaving the organization. We can also break down specific control points. So all these control points are layered on top of each other to develop a risk score for the transaction. But if there's a specific control point that we're seeing a lot of activity on that we want to individually investigate, we can now select that control point and separate that out from the other control points that were being triggered. So for example, if we wanted to understand, you know, the transactions that were triggering our outlier anomaly, we can go ahead and separate out for that particular area. So you can see how we can go ahead and filter and get very specific and what we're looking at to help monitor the organization in the type of transactions coming through. We can also build parameters for different types of users. For example, if a specific user 1796 only had approval, or only had the ability to enter transactions, but below a certain threshold, we can build out this filter for anything above that threshold and identify anything that went outside of the scope of her work his or her work and make sure that the proper approvals are taking place in the proper sign offs are taking place, once again, limiting that opportunity for that user to you know, perform a specific type of misappropriation of assets. So from here, we can go ahead and apply these filters and dig into those specific types of transactions. Once again, these are all based off of the risk scores here. So it's all being filtered off to that risk score. And another key piece of the program which helps take that risk based approach in that continuous monitoring service is around the intelligence sampler. So the intelligence sampler can be launched and we can work on a specific area such as expenses. And now we can go ahead and target all those high risk and medium rich transactions that are coming through, which are more uncommon and more unusual for this particular data set. Anything in low would be considered normal and common. So now we're able to target you know, and have a better approach to monitoring the activities within this particular entity. And any of these samples that would be selected here would then go into our audit plan which we could track and you know, follow up on and have a nice audit trail what's happening with this specific transaction that's been indicated as higher risk within this particular ledger.

 

Michael Bottala 

So as you can see,  MindBridge bridge can really play into the reducing the opportunity because of the visibility and the analysis that's occurring from the underlying transactions. So this type of service obviously helps prevent you know, behavior, Muybridge is not going to just capture those large, you know, material transactions, it's gonna be able to uncover a lot of those things. smaller items that, you know, may not be material to the financials. But as they accumulate years and years, you know, it ends up being a very large number. So, MindBridge believes that preventative behavior is going to lead to less big Miss big misstatements and big fraudulent activity in the long run. So that's really the core of what's happening around the general ledger analysis. Ken also mentioned we do have a payables module, which I'm going to dive into now. The very first thing that you'll notice about this payables module is it's very similar user interface to what you saw at the general ledger level, we try to maintain that similar intuitiveness of the different modules that we introduce. One of the key things that's different about the AP module versus a general ledger is we're getting very specific with the vendors that we're dealing with. This risk overview page can now be used to utilize or to break down the different vendors that we're dealing with. As well as the activity within those vendors. So just go ahead and do a quick refresh here.  Okay, there we go. So as you can see, we can select specific vendors and get a breakdown of the vendors activity over the course of the year as well as the type of transactions that are coming through and which control points are being impacted by this particular vendor. The way we like to analyze the vendors, we've introduced a couple control points which have similar concepts to what we discussed at the general ledger level, except now they're broken down by vendor. So instead of looking at the expense activity or expense flurries as a fund, we're looking at it for a particular vendor. So what that means is if we're issuing an invoice to a vendor on a monthly basis, and then all of a sudden we're seeing three or four invoices to that particular vendor. In the same month, that's gonna trigger an activity flurry, because that's outside of the typical behavior for that vendor. And we want to be able to surface that and look into, you know, why did this activity change? And why are we issuing more invoices in this particular month.

 

Michael Bottala 

Another control point that we explained earlier around the amounts that are being recorded to different funds and different general ledger accounts is now being applied to the vendors as well. So we're able to develop a neighborhood of what's normal for each vendor that you're dealing with. So if we paid a vendor 200 to $500 on a monthly basis, or per invoice, and then all of a sudden we're seeing a $10 invoice or $1,000 invoice, you know, those are outside of the norm in terms of the amounts that are being recorded. And now we want we want to bring those to the surface and understand why these amounts are different than what's typically been recorded. So those two control points along with us A lot of the business rules that we read, you saw at the general ledger level helps you analyze the activity within a specific vendor, an AP sub ledger. So if we go ahead and drill into a high risk transaction, again, you'll notice it's a very similar approach, we're able to see all the control points that are being triggered. For a specific entity. One of the things that we can do is we can build in some suspicious keywords. So for example, if there's particular vendors that are associated with family members within the the specific government entity, now we're able to build those in and flag those type of relationships very easily. Another thing we're able to do is there's typically an approved vendor listing, we're able to approve that compare that approved vendor listing to what's actually happening within the data. So if we're issuing any invoices to vendors that are not on that approved list, that's gonna be flagged by this control point right here. So there's a number of different ways that we're able to break down the information to really surface those unusual transactions to once again, prevent as much behavior prevent as much misstatements and fraudulent behavior as possible. The other key piece to the accounts payable area is how we're breaking down the vendor activity.

 

Michael Bottala 

So first of all, we can go ahead and look at it based off the ending balance at the end of the period, which is essentially the aging we're able to see you know what percentage of the total ending balance a particular vendor has, and what that change looks like year to year at at the end of the period. But with MindBridge since we do have all that underlying data, we're also able to break down the activity. So now we're able to get a good picture of any specific vendor that we've been dealing with over the course of the year and understanding how their activity has compared to prior years and as the activity as compared to the all the activity as a whole, so we can see if there's any concentrations within the vendors that we're dealing with. And we can see in this particular instance, there's been really these two related parties account for a lot of the percentage of the total activity. So that would indicate that these are, these are higher risk vendors that we'd want to look into because they're capturing a lot of the activity and they're related to somebody within the organization.

 

Michael Bottala 

The other key area of MindBridge is we incorporated a couple of the, you know, really key metrics within payables. One is days purchases outstanding. So now we're able to understand this specific ratio at a vendor level and look at looking at it year to year. So instead of taking this ratio as a whole, like more, most organizations will do, we can break down each vendor and that'll give us a better idea if there's any unusual behavior. happening within the transactions that are coming through. We also look at the AP turnover.  So the turnover gives us a good idea of you know how quickly these invoices are being paid to clients, and how that compares to the prior periods as well. I'm embracing Umbridge we really feel that visualizations also help tell a story about the data. So we can see we have these ratios here. But now we're able to see that on a on a graph over the course of the year. And by visually by visualizing, and seeing this on this graph, we're able to really pinpoint you know, the unusual behavior, we can see it's fall, oppose a pretty typical pattern compared to prior years, except now there's a spike in this particular month. So what's happening in that particular month now we can drill down into those transactions and get a better understanding what changed for these particular entities. If we move forward here into the data table, you'll notice that this is very similar to what you saw at the general ledger level, we can go ahead and parse the data, any which way to drill down into those transactions to get more assurance around any unique situations that we've identified. And if we have identified a unique situation, we can put in that same exact footprint into our filter. And we can identify all the transactions that follow that unusual circumstance, and quickly identify more more misstatements or fraudulent behavior within this particular entity. So as you can see, MindBridge is really, the goal of MindBridge is to provide more visibility to professionals such as Ken in order to make better judgments about what's happening and promote more preventative behavior within these organizations through services such as continuous monitoring, and like the Pun group offers. So that being said, I will go ahead and pass this back to you, Ken,

 

Kenneth Pun 

This is a really, really great demo here because one of the thing while we are looking at this, you know that most government agency, we don't have a whole lot of tools. And in this involvement by most of the government agency right now bad like currently dealing with is through the external audit and once a year, and the RS coming out to to perform the annual financial statement audit or compliance audit, and they just follow the rules and then using those traditional risk assessment methodology in identifying basically, it's, it's tough because you're not like, through the traditional external audit services, you will not be able to pinpoint like all the risks identified and then also with the traditional tools, you're not be able to evaluate what type of internal control because I have gone through our responsibility as the auditor as the external auditor out basically focusing on expressing our opinion whether the financial legal financial statement is presented fairly whether they're in accordance with generally accepted accounting principle here so I'm having said that like I'm getting like paying a little bit more like a lot more attention to the to the fraud area because even sometimes that you don't even know there's a big white elephant walking by and you have no idea because there are a lot of different cases a lot of different scandals in the past. Basically kind of like saying that hey, there are big thing flying by and and no one's actually catch it because one gumming up majority are pretty decentralized and trying to look at a detail of transactions. Which is really, it's really relying on the external audit so. So that's why the peering grip came into play. And I'll talk about like making, making gold transactions have visualized everybody and explained all all the risks that we can identify.

 

Kenneth Pun 

 Now we are able to explain that why this is different, and why this is a high risk. And so that like, I'm not saying that all transactions like coming out like weighted as high-risk is determined to be fraudulent, but it's actually worth taking some time and looking into the detail and trying to identify whether those transactions are legitimate or not. So utilizing our servers once we identify the risks we be like and then also speaking of like the value added service to The organization here is to work with the agency on control activities. Because right now as you can see, during this pandemic, everybody's on how like do actually realizing that there are cash shortfall in the last couple months and uncontainable couple of bars actually experiencing significant amount of cash  shortfall because they are a tourism city and in that situation that that that time they need to let go a lot of people and how does that happen? Like how is that going to help if you reduce the amount of people within your entity so that's why the backup added service about the transactions and as a service is looking at the risk assessment and then also helping the entity in the in designing and, and and mitigate some of the control because this is a very variable important insight like part of the service that we are providing information and communication that's a fourth element within the COSO internal control integrated framework, we are able to actually looking at the result from the AI technology and then have a visualized visualize platform dashboard to show you where the risks are. So that this is more proven to the to like we can communicate the result internally. And then also we can communicate this result to the governing bodies so that they can see where the risks are.

 

Kenneth Pun 

And last but not least, is monitoring the monitoring control is is the key. A lot of time you just use the light most of the government agency are using the budget as a budgetary control and compared and transactions but if you do actually use like if you're logged in using a what I call the Zero Based budget and just having inflation or deflation, so it's kind of hard for you to identify fraud. So once we get, like fraud, like on top of the anti fraud program, anti fraud control is all you need to proactively monitoring your data and proactively dealing with risk it's instead of just monitoring and without doing anything. So, in this situation, when we actually do provide the transaction set as a service to you, you will be able to like we are working with you closely and surrounding this internal control integrated framework ,framework as defined by COSO to address the risks that you are facing, especially in the procurement fraud risk here so there are a couple things as what what we said earlier, we do have a multi year analysis we do like journaling, journal entry from expenses to revenue. They're out like looking at the anonomys in general ledger so I'm not going to read it out loud on all the things that we can look at. But there's really an opportunity, I think the the control point really will be able to help us in translating and where the risks are and provide value added service to you so that you can strengthen your internal control and reduce the fraud risk there. This is really the last, last line on accounts payable looking at the  suspicious tender duplicate outlier amount, invoice entry control and split transactions.

 

Kenneth Pun 

Well, again, thank you so much for your valuable time spending with me and Michael this morning and looking forward to seeing you in the future. And if you do have any questions, feel free to contact us and we are more than happy to address your issues.