Why shouldn't we integrate directly with PayPal?

Hi there,

I think I might be able to offer some clarity here.

Firstly some disclosure: I used to work at eBay (UK), I like PayPal and have friends at both companies. I reckon PayPal will ultimately be a bigger component of eBay Inc than eBay. I now manage a company (http://vzaar.com) who has recently move too Spreedly this week. And then turned Spreedly off because of a PayPal issue (currently being resolved).

That said ...

I think what you need to do is what the two companies offer. As the person who made the business decision to go to Spreedly (against some much bigger competitors in this space) I can explain why I did so. I don’t actually see Spree and PayPal as competitors.

PayPal offers powerful payment transacting services.
Spreedly offers powerful business logic services for recurring payments.

These are very different things. In fact if you look at whats going on with PayPal this week (cf. http://x.com/ ) PayPal’s future plan is to get more and more OUT OF business logic and more and more into being the best transacting service with the most powerful API’s on the web. PayPal’s competition here is other CC gateways (authorize.net, braintree) and other subscription services (AWS dev pay, Google Checkout), not Spreedly.

What Spreedly offers is the business logic. You’re paying Spreedly a fee to do this for you. You’re paying them to think about the edge cases so you don’t. And boy are there edge cases.

You say all you have to do is code in some basic logic (accept details, upgrade, downgrade, etc etc) and you’re done. Which is true. Until you try do it, and you keep finding more and more issues. e.g.
- A user has a problem caused by you. You wish to refund them some time, or money? You need to build it in.
- A user wishes too upgrade part way through the month? Do you refund them the difference? Give them extra time?
- A user cancels their account. They’ve already paid though, so you need to cancel their account on their anniversary date. Then the user upgrades his account before it’s actually cancelled. How much money do you take from them? When is their new anniversary date.
- and so on and so on.

We’ve done this ourselves. And I’m sick of it. I spend half my time trying to figure out what to do in weird edge cases. Now Spreedly does it for me. And if another user on another Spreedly account finds a new edge case, and Spreedly fixes it and we get the benefits.

The second thing is that integrating PayPal is a really PITA. I spend development time trying to figure out bugs in the Payment system which I’d rather spend developing users features. Nathaniel can have that headache and I’m happy to pay him for it.

We spend 6 weeks implementing PayPal and the credit card side gave us headaches and we never turned it on. WE’ve spent probably several more weeks in the year fixing issues and headaches.

We spent a week implementing Spreedly.

The other side of the coin is that Spreedly doesn’t yet have all the features I want. But then neither does my system. I have a rough idea form Nathaniel when they are coming, and I know I couldn’t implement them faster than him anyway so I’m ahppy to wait. These include
- Recurring payments via PayPal accounts
- Overages
- PCI compliance.
- Customisation of the Spreedly collection pages

But whilst this is a (slight) negative it too is a positive. I know that Spreedly knows these are important for more than just us. Which means when they develop other features important to other customers too we both reap the rewards.

I am a big Spreedly advocate. Any system that takes none core stuff off my hands I love. It lets me focus on core stuff.

The same way I could do email quite easily I have no desire to run an email server. So we use gmail for apps.

You can implement Spreedly so quickly I would say the risk is low. If you spend a week and decide it’s not for you, you’ve not lost a lot of time. I guarantee the same can’t be said about PayPal, especially when you keep having to spend time on it throught the year.

I hope that helps you with your decision. This is pretty much the same case I presented to my board when explaining my decision to go with Spreedly.

I hope it helps you too.

Adrian

Adrian & Nathaniel: thanks both for your detailed replies. I think I kind of understand now.