There are multiple aspects to payment processing with basic payment processing and integration being the minimum -- Stripe currently has the most beautiful story here.
Things get more complex on the back end of the payment being processed at scale, namely fraud detection and risk mitigation. This is where some of those behemoths in the industry actually shine more.
Stripe passes the risk directly through, you are on your own.
For a lot of folks here working on startups with their first few-hundred customers, that isn't the problem -- the problem is accepting payments and getting back to work. I don't see anyone beating Stripe in that game.
For the folks that have huge customer bases, mitigating risk and dealing with fraud is a much more frequent/costly problem and having a processor help out there is why companies like PayPal (and BrainTree?) are still in people's vocabulary even though a good majority of their experience on the front-end suck.
This is one of those problems that doesn't matter to you at all, until it does... then it is a huge nightmare and you scramble around looking for a solution and suddenly realize why "everyone doesn't just use Stripe" -- or some equally hot new processing startup.
I'm not sure I agree with you. I've had to deal with plenty of fraud issues.
How is Stripe passing the risk on any more than other processors? [1] The end-result of fraud (at least, the kind I'm familiar with) are chargebacks. Those are going to fall back to you regardless of who your payment processor is.
We've had major fraud issues on our system which is powered by BrainTree. When you hit a certain chargeback rate, you basically get a call from the underwriting processor/merchant who says "Get your crap together or get lost".
My guess is that it depends on your business model.
(1) The risk to a TPPA model would be a fraudulent payment merchant signing up with your service buying from themselves with stolen cards. Charge-backs would occur here once the card owner discovered the card was stolen.
(2) The direct sale model risk would be selling and delivering a good or service and then the card holder (stolen or otherwise) charges-back to get the good/service for free.
In case (1) fraud detection would help detect the stolen card and mitigate the risk. In case (2), legitimate charge-backs from non-stolen cards are probably sometimes the bigger risk. I most likely didn't hit all the cases here but it certainly seems to be the case that PayPal or the like would not always help your business model with fraud prevention, supporting what callmeed said here.
Things get more complex on the back end of the payment being processed at scale, namely fraud detection and risk mitigation. This is where some of those behemoths in the industry actually shine more.
Stripe passes the risk directly through, you are on your own.
For a lot of folks here working on startups with their first few-hundred customers, that isn't the problem -- the problem is accepting payments and getting back to work. I don't see anyone beating Stripe in that game.
For the folks that have huge customer bases, mitigating risk and dealing with fraud is a much more frequent/costly problem and having a processor help out there is why companies like PayPal (and BrainTree?) are still in people's vocabulary even though a good majority of their experience on the front-end suck.
This is one of those problems that doesn't matter to you at all, until it does... then it is a huge nightmare and you scramble around looking for a solution and suddenly realize why "everyone doesn't just use Stripe" -- or some equally hot new processing startup.