As far as I can tell, "better".

Because they have their payments dataset, they're relying on having access to more non-public information to better train their models on.

The data is from Stripe's handling of 100,000+ businesses' transactions, probably a better dataset than SiftScience.

That's a big probably.

I would think that cuts both ways. SiftScience only needs to have a few big retail clients and they'll be doing a huge volume. More to the point I would think Stripe's strength here is their weakness....they're only as good as the users they have on Stripe vs Sift which might have a broader mix of datasets. Of course, as Stripe continues to grow that weakness will diminish.

doubtful that stripe has 100,000+ business customers. Kount is another good one, that's what Braintree uses apparently.

Why do you think it's doubtful?

They're processing around 100M API requests a day. https://twitter.com/patrickc/status/788752160284487680

From the page linked:

"We pinpoint fraud by building behavioral signals from across 100,000+ global companies."

Apparently they do ;)

Couldn't that refer to buyers of their customers vs the actual customer count?

Direct integration probably too... one-click to turn it on vs integrating w/ sift.

It's free, for starters...