The 2020 PwC Global Economic Crime and Fraud Survey found that 47% of the 5,000 companies they interviewed experienced fraud in the past 24 months resulting in a total loss of US$42bn. Fraud and economic crime do not only cause financial damage to businesses, they also affect local governments through the loss of tax revenue. As a result, in the last few years, governments have increased the number and severity of financial regulations such as AML.
For companies to comply with AML regulations, they are required to have a monitoring system to:
- Review Transactions.
- Calculate how potentially fraudulent a transaction is.
- Create risk alerts and compile them into cases.
An essential component to successful AML transaction monitoring is the quality of data that sales reps collect during the client onboarding process. It is crucial to AML monitoring as bad data can cause several problems for businesses that do not immediately take corrective action.
In this article, we answer how big of a challenge bad data is, and how it impacts regulatory compliance?
What is Bad Data?
Bad data is any information that can be flawed, misleading, and without structure. No one is immune to it, and if it is not acknowledged and fixed at an early stage, it can cause serious complications.
How does Bad Data occur – The Causes?
Traditionally companies will collect their consumer's data during the onboarding process and will continue to do so for CRM management purposes. CRM allows brands to manage their relationships with their customers to improve, retain and grow that relationship.
However, companies now are required to collect data as part of AML compliance and other financial regulatory compliance. Therefore, data collection has become more complicated, which means that to be able to fix dirty data, you must first understand the cause:
- Inaccurate data: Incorrectly entered or maintained information.
- Missing Data: There are empty fields with missing information.
- Poor data entry: Spelling mistakes, substitutions, and differences in spelling, naming, or formatting.
- Non-conforming data: Data which is not standard to the system of records.
- Inappropriate data: Submitting information in the wrong field.
- Duplicate data: Entering the same information under multiple fields.
What are some additional causes of bad data?
The first comes from the assumption that companies feed their consumer’s information directly into AML transaction systems. However, because the data is collected and maintained according to the different stages of the consumer life cycle, it rarely meets KYC requisites. Therefore, because the collection of information does not comply with AML/KYC requirements, the result is bad data.
Furthermore, because sales and support reps manually enter all the data relative to a transaction into their software before it reaches the AML Transaction Monitoring Systems, it can result in low-quality data. This process is known as enrichment, and it is highly sensitive to the smallest variation in data quality, i.e. due to human error, the slightest mistake in the data's entry will impact its relevancy and usefulness in regards to TMS.
Lastly, the fact that many businesses work with affiliate brands that target or benefit from the same customer base can complicate the collection of accurate and connected data even further. Meaning that the more parties that are part of the transaction process, the easier it is for inaccurate data to enter the AML monitoring system and the harder it is to make connections between data sets.
How does bad data affect AML compliance?
AML monitoring can only be successful when it has detailed info on the initiating entity, the business processing the transaction and on the parties who benefit from the exchange. The more quality details the system has, the more accurately it can detect fraudulent behaviour and high-risk customers. Therefore, when the info entering the database is inaccurate or irrelevant, it creates a 'Garbage in. Garbage out.' scenario that makes the monitoring process ineffective.
Low-quality data negatively affects AML compliance in the following ways:
- Results in an excessive amount of false positives
- Creates the need for additional due diligence
- Creates missing opportunities for identifying/detecting real hits
- Generating an excessive number of false positives
False positives are alerts for behaviour that seems fraudulent but isn't. According to the Global Risk Institute in larger entities, false-positive rates can be anywhere from 85% to 92% of all alerts generated. Each time this happens, an AML analyst must conduct further research to provide proof of why each highlighted activity is not fraudulent.
To investigate each alert, companies require additional workforce and resources, which means compliance teams must redirect their time and focus away from dealing with actual fraudulent cases.
Inability to comprehensively detect true positives
Even if companies employ the latest AI technology to monitor and alert them of risky customers, many AML systems will miss many true positives if the information in the database is inaccurate. And because employees are responsible for the data that enters a company’s database, they play a large part in low-quality data decreasing an AML programs ability to detect true positives.
According to PwC's survey, 43% of reported incidents come from internal sources and amount to US$100 million. Therefore, companies are not only losing money; they also face potential civil and criminal class actions, discredit their brand's reputation and damage their business relationships due to failure to comply with AML laws.
Creates the need for additional due diligence
In both cases of excessive false positives and failing to detect fraud entirely, the result is a need for additional due diligence. For example, every time the software identifies fraud the organisation's compliance team must (i) file a SAR and, (ii) go back and review the missed behaviour to retune the monitoring program to include that behaviour. Then they also must re-examine all the transactions that occurred within the last six months up to a year to see if any of the previous dealings from that time frame matched the same behavioural profile.
Cedar Rose Guiding Your Way Through AML Compliance
At Cedar Rose, we are your one-stop-shop for all your KYC and AML compliance needs. Our innovative database uses the latest in technology to ensure extremely accurate KYC due diligence reports. Our data warehouse currently contains identifying information on millions of companies and individuals.
Furthermore, we make sure to regularly clean our datasets so that we can promptly identify any incomplete, inaccurate, or irrelevant information and then either replace, modify, or delete it. Make smart business decisions by ordering one of our 15 KYC due diligence reports and receive instant verified results that rely on the highest quality of data.