1) I could accept payment into my own account and credit the users on my own system. I can then pay them when they request a withdrawal.
2) I could split each transaction into two parts, with my cut going into my account and the rest going directly to the user.
The first way seems the most reasonable and seems to be the way most sites do this, but I am scared of essentially storing other people's money in my account (especially after reading stories of Paypal accounts being frozen). I am also unclear about the legal issues of doing either method.
Any tips on how to proceed? Any payment processor recommendations? I am a poor college student, so cost of setup is a major issue for me. Thanks!
Edit: the site is a marketplace, not some type of money-transferring service
You would have a very difficult time establishing a proper merchant account as the fraud risks are incredibly high. While many might be willing to discuss it, it is unlikely that the account would be approved by the processor's sponsoring bank in the underwriting phase.
Amazon's FPS service, while far from ideal in that it requires all parties have Amazon accounts, and requires the transaction occur off-site in Amazon's payflow, is very well-suited to this use case. It's the only quick, affordable, easy to integrate service I can think of that allows for a merchant to broker transactions between two parties (while skimming a percentage).
If you push forward with this, I'd definitely recommend launching with FPS rather than plunking down a large personal guarantee with Authorize.net or Braintree.
Good luck!
Quick summary of my experience:
1. Billing credit cards is easy, paying people through an API using direct deposit is pretty hard to find
2. Lots of companies won't know what you're talking about. The ones who do will say you are an "aggregator" and high risk.
3. BrainTree payment solution is a possible exception to this. They "get it" but they also have rejected everyone I know who has applied. They rejected my app also. You need a mil in sales and a track record.
4. I finally stumbled upon AchDirect.com. They are a two-bit no-name processor in Texas, but they approved my app, and give you an API to do direct deposit. Their customer service was incompetent, documentation was sub-bar, and generally didn't seem like they knew what they were doing. But i did finally get the thing to work.
I had to write a custom ActiveMerchant plugin for AchDirect (if you want the code get in touch with me).
I think WePay is using AchDirect too (not sure).
Anyway, Amazon FPS is much simpler. But your customers experience is much worse. It confuses people because they don't think of Amazon like they do Paypal. And (at least when I tested it) there were too many screens to go through to complete a simple transaction.
Anyway, that was my experience.
Thanks for pointing me to Amazon though. I'll look into it some more.
Paypal is definitely very easy to set up, but I keep hearing stories of accounts getting frozen. If I do opt to use it to hold the earnings of sellers, and that account got frozen, it could be disastrous. The second way would be far less risky because it would just be the site's cut frozen, but splitting up payments doesn't seem to flow well with Paypal (maybe I'm wrong about this?).
In terms of revenue. It's always helpful if a business already has a track record and some external funding for validation as well as some stability. That also clearly presents a catch-22. Some providers are wiling to take substantial risks: new business, peer-to-peer track record and no funding while others are only willing to consider this type of high risk businesses who have an established track record.
I would offer a friendly word of caution. All too often new companies will find a provider to approve this type of business model only to be shut down at the worst possible time. This happens because sometimes the sales rep doesn't properly explain the business model to the underwriters, the underwriters don't understand it, and/or an application gets auto approval because it's new and has low expected volumes. I'm not suggesting that there are not providers that won't approve this type of business, and I'm also not suggesting that everyone approved for this type of payments model will eventually get shut down. What I am suggesting is that we see merchants getting shut down too frequently, so I'd be cautious if I were you.
Best of luck to you.
Aggregation more fully explained: http://www.braintreepaymentsolutions.com/blog/high-risk-mech... More information about payments and risk: http://www.braintreepaymentsolutions.com/services/new-to-pay...
Also in the long run you'll have to support more than few modes of withdrawal: checks, payoneer, moneybookers etc. PayPal is still very popular and your users will eventually demand it. If you have international users then it adds another layer of complexity.
