This is a massive win. I remember trying to setup payments and begin rejected several times by Worldpay and this was a process that took several months and required us to keep funds in escrow. Ultimately we settled on PayPal integration which required sending customers off to paypal.com and hoping they went through with it.
I'd love to see some case-studies in terms of conversion rates for completing transactions when it comes to keeping customers on site to do payments versus sending them to a 3rd party.
The blog post was discussing the 2008 version of the Braintree APIs. We completely rewrote our core APIs in 2009. Today, I think we provide one of the most elegant payment APIs available.
We've also significantly decreased the amount of time it takes to get a merchant account -- we have an instant application process now in the US, so many of our merchants are approved and provisioned with an account as soon as they hit the "Apply" button. Our international application process isn't instant yet, but we're working hard to make it as fast as possible.
And the Braintree API is almost as good as Stripe's, if not equally good (it lagged a couple years ago, but they've been catching up). Stripe is a little bit slicker in my view, but not dramatically so. For normal stuff, your'e not going to see a huge difference in development time - certainly less than a man-month with either provider.
I eventually switched to Stripe because it was cheaper at low volumes (this was important to my project since it was bootstrapped), simpler, and they never asked for a personal guarantee. But it wasn't a huge difference operationally. Both companies provide very good support and are perfectly acceptable.
Both are miles ahead of Auth.net and Paypal, which are pretty horrible.
Honest question: What is it about Stripe (picking this one specifically due to having used it) that would take a month of work? Their documentation is thorough but -- at least last time I used it -- totally lacking in examples, so patching together a server-side workflow for creating and charging against profiles took a couple days of on-and-off work, but a month? I can see this if it includes other development on the project, but as far as actually getting the relevant parts of the API integrated and working, I'm wondering if other people's experience has been vastly different with this.
Where these new services win out is the ability to instantly be underwritten as a merchant and start processing. For WorldPay I had to submit passport, bank account statements (via fax no less), sign my life away etc.
http://www.kalzumeus.com/2012/08/06/stripe-and-ab-testing-ma...
I wonder if Balanced has good integration with the third-party shopping carts out there...
There's much higher fraud risk with online card payments than in-person ones. There's both risk that the customer presenting the card number isn't the cardholder, and risk that the merchant is itself a fraudster trying to charge stolen cards, launder money, or accomplish some other illegal act on the network.
Traditional merchant accounts mitigate some of that risk through an extensive underwriting process. Stripe does not have such a process. The only way they can mitigate the risk is by identifying patterns in transaction activity and getting chargebacks/disputes from cardholders. The 7 day period gives them sufficient time to identify the patterns and receive early disputes before the money leaves their control.
They also have an API but I have not really looked into it.
Balanced abstracts away this batching process and giving you events and nice json objects to deal with rather than parsing the batch file that eventually gets returned.
You could make the same argument for the card processing, if you really really wanted to you could bypass Balanced/WePay/Braintree/whoever and deal directly with the bank but for most people your time is better spent getting on with your core business.
The security conscious part of me also likes that there's human, offline intervention before money leaves my bank account.
Ok, I get you want to be hip and all, but Mollie will someday apply for like a real job in the real world, and the company will look at her resume and go "Director of Happiness...very pretentious, are the rest of us Directors of Misery or what...", and that resume meets the shredder.
Why not give Mollie a real, you know, non-vacuous title ?
I myself go back and forth between "Lead Janitor" and "Co-founder".
I was thinking escrow for my startup originally but stripe apps offer a different approach - i.e. cash goes straight from tenant to customer with fees (including a handling fee paid to you) deducted directly from the tenant.
It's a nice alternative to a straight-up escrow setup.
Many marketplaces want to avoid the liability for transactions that take place on their platform. This is specifically why we built Stripe Connect (https://stripe.com/connect). It allows the marketplace to provide a way to accept payments to their sellers. The buyers and sellers can the transact with each other and the marketplace avoids liability.
Balanced provides the ability to pass disputes onto the seller, but the marketplace ultimately has full liability for chargebacks, fraud, and so on.
With Stripe Connect, each seller is responsible for charges run through their own Stripe account. This removes any liability from the marketplace, and is really useful if you want to purely be a platform of facilitating payments and don’t actually want to be involved in the risk and complexity of handling any of the money.
Example: Buyer pays $10, seller gets $9, marketplace gets $1.
A lot of the successful marketplaces these days (e.g. Airbnb) have decided to fully intermediate the payments on their platform for security reasons. They find escrow is a way to promote trust and security on their platform. Most third-party payment platforms don't allow for this. Even eBay, which uses the most popular 3rd party payments platform of all (PayPal) had to initiate a $5000 "PayPayl guarantee" to encourage sales. Other marketplaces, like Etsy, are moving away from PayPal and building their own intermediated payments system to have a similar experience as Airbnb. My understanding is that Stripe Connect is great for store builders like Shopify, which are disaggregated, but less attractive to marketplaces like Airbnb.
Stripe quite simply wasn't the service they needed and they made a mistake in initial selection. This article title implies that Balanced is "better" than Stripe when they are quite different services entirely.
Is Balanced better than NYT?
Stripe just launched in beta in the UK and is available in Canada as well, whereas I believe Balanced is only available to users in the U.S.
Many customers with international buyers and sellers have opted to use our Stripe Connect product for that reason: https://stripe.com/connect
There are a few major differences between Stripe Connect and Balanced. Amber from our team details them here: http://www.quora.com/Balanced/How-does-Balanced-compare-agai...
I really like the layout of the blog. It's nice looking and clean.
The other comments about choosing the wrong tool are apt, but it's also true that this is a part of the payments business. Stripe can choose to do the sorts of holds and escrows these folks are looking for, or not, but I do think Stripe is particularly easy to integrate.
I think this piece is really trying to articulate the grouptalent business model to developers and it comes off pretty well to my mind.
Aside from that, paying two parties for every contract introduces friction. Multiple payments means more work, more manager approvals, more reimbursements for the company card, etc. Some portion of the customers would simply choose not to use the service because of that added complexity; it'd be lost business.
If Balanced is that easy and enables escrow-style payments, that's pretty brilliant. Will definitely be looking into it for future projects.
How and why are you willing to float that risk?
