vault cards redundantly

[chadwhitacre]

We can't afford to lose cards again. Can we vault with SpreedlyCore and Balanced?

Need to figure out both information flow and how Gittip would include this in pricing.

[justinbenson]

Hey there. Justin from SpreedlyCore. Thanks for thinking of us. I think Balanced looks very interesting and I'm glad they got involved. Core could store the cards for you. We also plan to support non credit card transactions like Dwolla very soon (week or two) and more after that with Core so could help there. Maybe the Balanced folks have some thoughts on how it could work together.

[chadwhitacre]

@justinbenson Thanks for chiming in. What's the normal data flow? I want to store my users' credit card numbers in two places, not just one.

[justinbenson]

The normal data flow is just one. You write to the Core API, drop in your gateway credentials, we store the cards at Core and therefore you have gateway independence. I'm not sure what BalancedPayments does - if they too store the cards away from the gateway and vault them separately. If yes, and they were up for it, there might be a world where they integrate to Core and create a duplicate storage scenario. Not sure how they'd feel about that.

[chadwhitacre]

Ping @mtamizi @mahmoudimus @jkwade. Do we have any options here? From Gittip's pov I believe there is value in vaulting cards redundantly. Open to discussion and suggestions ...

[matin]

@whit537 I apologize for the delays the last two days.

I'm happy to come up with a creative solution, but could you outline the use case?

Samurai is an anomaly. Be it Balanced, Stripe, Braintree, or Spreedly, we all have practices on migrating card information amongst each other or to any other PCI certified provider. Migrating the cards from Stripe to Balanced was a clean process:

1. Gittip initiated the process
2. Stripe verified Balanced is PCI compliant
3. Stripe encrypted the card information using a pub key provided by Balanced
4. Balanced loaded the cards
5. Balanced provides a mapping of Stripe to Balanced card tokens to Gittip. (@mjallday should have already email you about this)

[chadwhitacre]

I'm going to be on vacation this week. I'll be back on Friday, August 10 to run Gittip #10, and will pick up with development the week following. In an emergency please call my cell, +1-412-925-4220. Thanks!

[chadwhitacre]

This stalled. We are in your hands, @matin. :-)

[chadwhitacre]

Reopening in light of Balanced shutting down.

[chadwhitacre]

I've read https://spreedly.com/files/spreedly-business-case.pdf and have created an account.

[chadwhitacre]

Looking at pricing, we can fit in the $99/mo plan to start. That gives us 5,000 vaulted cards and 5,000 transactions per month. We have 3,052 working cards on file, and we do about 4,750 charges per month. Pricing increases linearly in $100 increments up to 45,000 cards and transactions per month.

[chadwhitacre]

I've set up a test environment on Spreedly.

[chadwhitacre]

Looks like there is no official Python library, though are several unofficial ones:

https://github.com/mooballit/spreedly-core-python
https://github.com/ridgebit/python-spreedly
https://github.com/shelfworthy/python-spreedly

[chadwhitacre]

I've completed their quick start tutorial. There's a few quirks but all in all my first impression is that it's a solid API.

[chadwhitacre]

How strong is their business? How much investment have they taken and who are they competing with?

[chadwhitacre]

https://www.crunchbase.com/organization/spreedly

Funding Received: $2.2 Million in 4 Rounds from 1 Investor
Most Recent Funding: $899.7 Thousand Seed on January 22, 2015

http://blog.spreedly.com/2015/01/06/crossing-50-million-per-month/

At Spreedly we’re challenging the status quo that a single payment stack is the right way to scale for web and mobile payments. December 2014 was our 22nd month since we (pivoted) launched Spreedly as a service that solves for the development and data ownership/PCI challenges in digital payments and maximizes the ease with where and how you transact with your data. As the blog post title states our customers processed $50 million in credit card transactions on our platform for the month of December 2014.

Many of our earliest customers took a big gamble on relying on such a small startup to help power their payments. So we want to let you know that we’re making great progress and plan to be here for a long time in the future. And better still if you’re one of those who’s been on the fence come on in the water’s fine.

[justinbenson]

A little over $2 million. We don't have any direct competition. Our
competition is say Stripe or Braintree where platforms decide whether they
want to work with one single processor or have support and flexibility for
many.

Our growth is very good/solid. We're a SaaS business so that makes life
more predictable.
On Apr 2, 2015 9:41 AM, "Chad Whitacre" [email protected] wrote:

How strong is their business? How much investment have they taken and who
are they competing with?

—
Reply to this email directly or view it on GitHub
#210 (comment)
.

[chadwhitacre]

From the FAQ on the pricing page:

What happens if my usage exceeds the plan that I am currently on?
Your service will not be interrupted. We will automatically update you to the appropriate plan for your usage.

[chadwhitacre]

@justinbenson What's your biggest challenge right now as a company?

[chadwhitacre]

Here's their GitHub: https://github.com/spreedly.

[justinbenson]

Awareness. We're coming at payments differently. Just focused on APIs and
PCI and not trying to be the processor/independence there for our clients.
So people don't know how to view us.
On Apr 2, 2015 9:53 AM, "Chad Whitacre" [email protected] wrote:

@justinbenson https://github.com/justinbenson What's your biggest
challenge right now as a company?

—
Reply to this email directly or view it on GitHub
#210 (comment)
.

[chadwhitacre]

Having had to move gateways three times now, your value prop is clear to us. One thing I'm not seeing is pay_ins_ pay_outs_. We're a marketplace and need to do ACH debits credits in addition to charging cards. Do you do that and I'm missing it? If not, is it on the radar?

[chadwhitacre]

Dang, sorry ... I'm not seeing pay_outs_. ;-)

[chadwhitacre]

https://twitter.com/whit537/status/583611866703728640

[chadwhitacre]

Aaaaaand s/ACH debits/ACH credits sorry. 🐑

[justinbenson]

We just support echecks through a few gateways today. Bigger plans but late
summer so a few months away for sure.
On Apr 2, 2015 10:11 AM, "Chad Whitacre" [email protected] wrote:

Aaaaaand I meant s/ACH debits/ACH credits sorry. [image: 🐑]

—
Reply to this email directly or view it on GitHub
#210 (comment)
.

[chadwhitacre]

Okay. eChecks are still payin so I guess we solve payouts separately for now.

[chadwhitacre]

This is basically an insurance policy on our card numbers.

[chadwhitacre]

This is an expensive insurance policy. Our current highest expense is Heroku at $90/mo (since we pass through processing fees to our users, those costs aren't equally visible to us). Cheapest option here is $99/mo. This also introduces friction, another moving part. Do we want to use the Braintree API, or the Spreedly API? We're under the gun on #3377 and I'm afraid that more moving parts means more implementation risk, slowing us down. :-(

[chadwhitacre]

This also introduces friction, another moving part.

E.g., on https://docs.spreedly.com/payment-gateways/braintree/ I'm seeing credit card and Apply Pay as supported payment types. What about PayPal, Venmo, and Coinbase? If we work with Braintree directly we don't have the impedance mismatch.

Also, what's up with the "orange/blue" distinction? Where does that come from? I haven't seen that anywhere on Braintree yet. Is that a Spreedly distinction? What does it mean? Why?

[chadwhitacre]

Braintree's vault (when using the "blue" API) has a top-level concept of a Customer, and Core allows limited interaction with the Customer when storing credit cards in Braintree's vault.

Another example of friction. What if we need to do more with the Customer than Spreedly allows for? This makes me nervous.

[chadwhitacre]

Making the call on this. Deciding Spreedly isn't the right move for us right now, because of cost vs. benefit, and increased implementation risk on #3377.