Let’s play - Sift’s FIBR tool

Have you already checked Sift’s FIBR tool? I’ve played a bit with it to see which industry struggles the most with fraud. Unsurprisingly, it turned out that aggregation platforms are getting the brunt of it - PSPs, travel portals & marketplaces most of all.

But first, a bit more about Sift’s publication: the Fraud Industry Benchmarking Resource (FIBR) was published just a couple of weeks ago, and it’s worth to mention a few disclaimers before talking about insights:

• Sift didn’t really share the methodology behind the metrics, and anyone who ever reported on financial fraud knows how different results can be based on the measuring definitions.

• It goes without saying that Sift is not an NGO. They have an intrinsic motivation to look good, making FIBR a great resource for outsiders as well as a great marketing tool. Still, kudos to them for revealing any numbers at all, that’s much more than most vendors do.

• For crying out loud, why not call it the “Fraud Benchmarks Index”?! It was there for the taking!

Disclaimers aside, here are the worst performing industries worldwide by the measured metrics:

Payment Fraud Attack Rate:

• 1st place - Travel

• 2nd place - Payment Service Providers

Fraudulent Chargeback Rate:

• 1st place - Ticketing

• 2nd place - Marketplaces

Manual Review Rate:

• 1st place - Transportation

• 2nd place - Marketplaces (ignoring the Loyalty category which has an astounding ~20% review rate, which seems to me a bit too skewed)

Why didn’t it come as a surprise to me? Well, there are a few factors that make it harder for these industries to perform well. First, they attract more fraudsters. Not only because they have a higher brand awareness, but mainly because they usually have a much wider “surface” to attack. For example, by offering more payment options and/or servicing more markets.

They also lack intimate familiarity with the sellers and products on their platform. The sheer scale of their offering comes at the expense of identifying different fraud patterns, exploiting niche data sources and correctly classifying legitimate buyers that behave unusually.

Lastly, these platforms lack standardization as they are mostly developed from scratch both on the tech and the operational levels. In turn, this makes it harder for vendors to roll out products that perform well “out of the box” for these segments.

Looking ahead, the picture doesn't get any better. Such aggregators are usually not big enough to buy an existing fraud solution and at the same time don’t have the profit margin to heavily invest internally. As I don’t expect these dynamics to change soon, I believe they will continue to struggle in the foreseeable future.