Eight Reasons Why Search Engines Are Not Due Diligence

2018-October-11

Search engines are free, easy to use and powerful. Could you use them to run a KYC/AML screening program? You could – but in the end, it will expose you to significantly more cost and risk than opting for a purpose-build solution. Here are eight reasons why.

1. Data from Search Engines is Unstructured

Search engines return stories and URL links, when what you ideally need are consolidated profiles, with secondary identifiers such as age, location and connections organized for easy review. Sometimes, you’ll need content from multiple stories to make an onboarding decision.

2. Search is Inefficient

First, you get pages and pages of results, many of which are not relevant. Second, there’s a lot of duplication, since high-priority stories are often syndicated and therefore repeated many times. This creates huge volumes of false-positives that analysts must research. Finally, search engines do not allow for daily or periodic review of adverse media, creating even more work for analyst teams. The result is tremendous inefficiency and excessive compliance costs.

3. Search Engines are Biased

Search Engine Optimization (SEO) can manipulate the results that are returned in a single search, pushing certain topics to the top of a search or others to the bottom. Search results are also influenced by users’ locations, what they have searched for previously. This means inconsistent review across different analysts and makes it much harder to identify risks efficiently.

4. Important Intelligence is Missing

As publishers move content behind paywalls/subscription controls, it’s only indexed on search engines, which don’t provide the full reference or story. Content is also sometimes archived and removed from search indexes over time. All of this can result in ‘missing’, risk-relevant content that may be critical to due diligence decisions.

5. A Search Engine’s Priorities are Not Your Priorities

Search engines prioritize news based on the popularity of the story or source. This has no relevance to your rules and priorities around due diligence.

6. How Do You Know it’s Not Fake News?

Search algorithms are confidential. But they all essentially prioritize news based on clicks and indexing, not on the integrity of sources. For adverse media, this means that stories from anonymous or unknown authors or people of questionable integrity are placed alongside valid sources. Again, this makes it harder to find truly risk-relevant information.

7. You Can’t Audit a Search Engine

Because search results are constantly updated, the same search carried out just minutes later can return different results. This creates complications for quality control, regulatory reviews, and decision-making. There is also no way to track a profile at a point in time and store those results for future audit needs.

8. Monitoring Becomes Your Problem

Due diligence doesn’t end when business relationships begin. Any customer or partner could become involved in activities that conflict with your compliance policies and rules, putting your organization at risk. Organizations relying on search engines are forced to screen either once a year or take a segmentation approach, only re-screening the highest-risk individuals. Both these methods are inefficient and can leave you exposed.

So what’s the answer?

Don’t rely on search engines for due diligence. RDC’s solution, for example, overcomes all of their limitations. It is:

Structured – we aggregate, curate and index data, and stored as profiles for efficient searching and filtering

Efficient– machines do the hard work, with AI-based name matching reducing false positives by up to 90%

Unbiased – consistent filtering and prioritization algorithms eliminate subjectivity and search engine biases.

Complete – we collect data from more than 200,000 reputable sources, with a full audit trail

Prioritized – we remove “popularity bias” to focus exclusively on risk-relevance

Credible – “fake news” is eliminated using source validation, NLP technology and expert curation

Auditable – RDC’s data and screening platform delivers consistent, documented and auditable results

Monitored – your portfolio can be automatically monitored every day with no extra work for compliance staff.

 

18

September

2018

It Takes Two To Tango: The Interdependence of Data and Technology

By Michael Kerman | September 18, 2018 Everyone’s talking about artificial intelligence and machine learning (AI and ML). But what’s […]

View more
11

September

2018

Combating Modern Slavery and Human Trafficking: What Financial Institutions Should Know

The issue of modern slavery continues to be a sensitive topic within the private sector. A lack of experience of […]

View more
03

August

2018

Using Cognitive Prediction to Confront Terrorist Financing

Traditionally, the approach to dealing with terrorist financing starts by breaking it down into two kinds of needs: supporting organizational […]

View more
24

July

2018

Tracking Risks Regarding Marijuana Related Businesses: Seed to Sale Monitoring

With conflicting opinions from state and federal legislation and limited guidance from the Department of Justice and FinCEN, financial institutions […]

View more

Best in Class Onboarding

Sarah Kocianski: Best in Class Onboarding | June 12, 2018 (A version of this article was originally published on 11:FS […]

View more
26

June

2018

Insurers turn to artificial intelligence in war on fraud

By Steven Melendez | June 26, 2018 (A version of this article was originally published on FastCompany.com) Using artificial intelligence […]

View more
26

June

2018

Modern Slavery Typologies for Financial Services Providers

by The Mekong Club   | June 26, 2018 This project aims to equip financial service professionals with an understanding of […]

View more
19

June

2018

Surfing Uncertainty To Confront Terrorism Financing

by Thomas M. Obermaier, RDC   | June 19, 2018 Keynote presentation to the FinTech FinCrime Exchange (FFE) in London, UK […]

View more