Case study

Insolvency firm Griffins speeds up fraud forensics with IBM analytics

Brian McKenna

Insolvency firm Griffins is using analytics software to reduce the cost, time and complexity associated with fraud and forensic investigations for litigation.

Griffins has a team of around 50 investigators, covering both insolvency and financial investigation in the UK.

fraud.jpg

The firm has been using IBM i2 Intelligence Analysis software for three years. Stephen Hunt, insolvency practitioner and partner, says the firm has learned from police and government use of the technology, and taken it beyond its use as a reporting tool for court – what he calls the “analyst notebook” element.

Hunt recalls that he was a “little dubious as to its value originally, but I saw how it could be adapted to our own purposes”, to help with pattern analysis. “You can write huge structured queries to suit any forensic question," he says.

Griffins' investigators glean critical insights when investigating complex incidents, through the software’s visualisation of people and events. It is also used to document results for potential litigation.

The IBM i2 iBase database application has relieved Griffins from struggling with more than 10TB of data stored in Excel spreadsheets to “a faster, more accurate, intelligence-led approach that helps solve cases related to money laundering, missing trader fraud, and theft of company assets”.

Looking for ways to recover money is in vogue, and software is part of that

When tracing the proceeds of crime, the firm uses the software to analyse structured and unstructured data sources, such as bank statements, PDF files, emails, invoices and spreadsheets. Through this, it establishes patterns and relationships, making non-obvious connections between disparate sources of data.

Hunt gives the police analogy of “consequential transactions”, where, as an example, a person makes a telephone call, then makes a payment immediately, on a regular basis; or when a mobile telephone call is made near a specific cash machine, time and again. 

“You can take separate data stores and put them together,” he says.

Griffins is applying that approach “to dull accountancy packages”, in tandem with telephone records, bank transactions, and so on, related to a company under investigation.

The firm has recently imaged a large amount of data from a bank in the Caribbean with a largely fraudulent turnover of $1,000bn. From the partial data available to them for civil claims, they were able to “extract highly accurate patterns and make connections during investigative analysis” which led HM Revenue & Customs to query how they had obtained their information, he says.

“It would take months to do that evidentially - it can be done in minutes with our database. That has been revolutionary”.

Hunt believes this type of software could develop into a problem-solving package of algorithms that will answer the top five questions that often go un-posed in insolvency investigations, due to cost.

“You could apply [those algorithms] to an accounting package, stand back and it will do the investigation for you” – at least terms of seeing where fraud is, or is likely to be.

“Imagine where a firm goes bust owing millions of pounds, with no assets when it goes into liquidation. Who will pay for the investigation of that? The creditors?” And so technology of this kind can help insolvency practitioners get quick leads, he says.

“Fraud is ever present and is far more widespread than you can possibly imagine," he says. "And there is more focus now on it at a governmental level, in terms of tax fraud. Certainly, looking for ways to recover money is in vogue, and software is part of that."


Email Alerts

Register now to receive ComputerWeekly.com IT-related news, guides and more, delivered to your inbox.
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
 

COMMENTS powered by Disqus  //  Commenting policy