payments and multi-vendor e-commerce

9 views
Skip to first unread message

bobhaugen

unread,
May 19, 2009, 9:57:03 AM5/19/09
to pinax-business
I am embarrassed to say I have never worked on the payment modules of
an e-commerce system. Somebody else always did it.

But I have talked to one of those somebodies, and she says there are 4
main pieces that an e-commerce vendor needs to accept credit card
payments:

1. A credit card gateway, for example Authorize.net or Skipjack.

2. A processing layer between the gateway and the bank, for example
http://advancedpaymentsolutions.net/

3. A merchant account at a bank.

4. An SSL certificate.

That's a lot of setup. Each piece has subscription costs, and several
of them also have transaction fees.

So I'm wondering if a multi-vendor social e-commerce system could use
one set of all of those components for all of the member vendors?

And what other complications would follow, like automated wire
transfers to banks to get the money to the vendors?

Anybody got enough experience with these matters to hazard some
guesses? Or even better than guesses? Like, this is a really stupid
idea, or it has half a chance?

sste...@gmail.com

unread,
May 19, 2009, 9:59:46 AM5/19/09
to pinax-b...@googlegroups.com

On May 19, 2009, at 9:57 AM, bobhaugen wrote:

>
> I am embarrassed to say I have never worked on the payment modules of
> an e-commerce system. Somebody else always did it.
>
> But I have talked to one of those somebodies, and she says there are 4
> main pieces that an e-commerce vendor needs to accept credit card
> payments:
>
> 1. A credit card gateway, for example Authorize.net or Skipjack.
>
> 2. A processing layer between the gateway and the bank, for example
> http://advancedpaymentsolutions.net/
>
> 3. A merchant account at a bank.
>
> 4. An SSL certificate.
>
> That's a lot of setup. Each piece has subscription costs, and several
> of them also have transaction fees.

Or, you could just go through PayPal or Google's checkout system and
save all of those troubles if cost is such a significant issue for
this particular set of users.

S


bobhaugen

unread,
May 19, 2009, 10:18:54 AM5/19/09
to pinax-business
On May 19, 8:59 am, "sstein...@gmail.com" <sstein...@gmail.com> wrote:
> Or, you could just go through PayPal or Google's checkout system and
> save all of those troubles if cost is such a significant issue for
> this particular set of users.

It's not just the cost, it's all the setup. I'd prefer closer-to-
instant gratification.

I'm aware of those services. And maybe having the e-commerce software
try to handle it all is a stupid idea?

If I understand correctly, both PayPal and Google Checkout require the
merchant to have an account.

Google Checkout appears to require the buyer to have or set up an
account, too, while PayPal apparently does not.

Just learning how to think about this stuff, and all the associated
trade-offs.

sste...@gmail.com

unread,
May 19, 2009, 10:28:11 AM5/19/09
to pinax-b...@googlegroups.com

On May 19, 2009, at 10:18 AM, bobhaugen wrote:

>
> On May 19, 8:59 am, "sstein...@gmail.com" <sstein...@gmail.com> wrote:
>> Or, you could just go through PayPal or Google's checkout system and
>> save all of those troubles if cost is such a significant issue for
>> this particular set of users.
>
> It's not just the cost, it's all the setup. I'd prefer closer-to-
> instant gratification.

Then you have to figure out how to multiplex a single holding account
and disburse money to the correct customer. Huge overhead and more
cost and setup by far than setting up a simple PayPal account.

> I'm aware of those services. And maybe having the e-commerce software
> try to handle it all is a stupid idea?

Yup.

> If I understand correctly, both PayPal and Google Checkout require the
> merchant to have an account.

Or, see above; you have to disburse the money at some point making you
responsible for all the bookkeeping, accounting, taxes, etc. Bleh!

> Google Checkout appears to require the buyer to have or set up an
> account, too, while PayPal apparently does not.

Right. PayPal's the lowest impedance but some customers resist using
it if they see it, which they don't have to.

S

bobhaugen

unread,
May 19, 2009, 10:36:41 AM5/19/09
to pinax-business
On May 19, 9:28 am, "sstein...@gmail.com" <sstein...@gmail.com> wrote:
> Then you have to figure out how to multiplex a single holding account
> and disburse money to the correct customer. Huge overhead and more
> cost and setup by far than setting up a simple PayPal account.
>
> Or, see above; you have to disburse the money at some point making you
> responsible for all the bookkeeping, accounting, taxes, etc. Bleh!

I already got code that does all the disbursements, bookkeeping, and
accounting. Need it where one customer order contains items from
several vendors. And in some cases, the administration takes a cut.

So it's sortof a natural to think of handling the payments for
multiple vendors with one gateway, at least.

Social Network in DJango

unread,
May 20, 2009, 9:32:16 AM5/20/09
to pinax-business
I would find a model that integrated with the vendor choice of paypal
and google checkout, to be the most
useful to our project and probably many similar projects.

Our revenue model is slated to be a monthly membership fee to access a
social website. There would also be digital content for sale.

The core issue in setting up an e-commerce situation is building
trust, particularly with customers. Using an existing proven end-to-
end system
is much more trustworthy in their eyes than even a far technically
superior low-cost system that we implement as a team or as
individuals.

I am guessing that a django reusable application that integrates with
such payment services could be written by someone with experience over
the course of a number of weekends.

I applaud Bob for pushing this whole concept forward.

Jonathan
p.s. Bob, would you consider offering your accounting software as a
project in pinax? I will need to be tracking micro payments.

bobhaugen

unread,
May 20, 2009, 10:56:55 AM5/20/09
to pinax-business
On May 20, 8:32 am, Social Network in DJango <readb...@gmail.com>
wrote:
> p.s. Bob, would you consider offering your accounting software as a
> project in pinax? I will need to be tracking micro payments.

My code will end up being open source.

My goal is to develop a full suite of business apps that work with
Pinax. But my process is to develop a series of one-off projects for
particular user groups and then generalize from there. In other
words, from the particular to the general.

Or in still other words, the first iterations will be tailored to
their particular situations and may not work for yours, but you will
be welcome to mine code.

Two early iterations are the swaps app in current Pinax, and
http://code.google.com/p/pinax-cohousing/

I'm working on another system for a food distribution group in
Wisconsin, that I will publish when it's a little more stable.

None of those systems do consumer e-commerce. Swaps is just swapping
stuff, cohousing just does organizational stuff at this stage.

The food distribution software is second-generation. The first
generation was my first Django project, and I would be embarrassed to
open-source it. The second generation should be better, but still not
very general at all. And it just does the back-office business
stuff. Their customers do not want to enter orders, they want to talk
to live people.

Cohousing is Pinax, the food distribution stuff is not.

Talking to two different groups now about new social business projects
that will be Pinax, and will do consumer e-commerce.

Hope that all made sense.

Social Network in DJango

unread,
May 21, 2009, 1:13:01 PM5/21/09
to pinax-business
Hey Bob,

I suggest two concrete use cases (not necessarily what I'm doing).

(1) I heard of a twitter integration called tipjoy that claims to
facilitate micro payments with a command line interface from a social
network.
As I understand it, each contributor signs up with an account through
paypal, putting at least $5 in their account.
When they wish to spend money on the system, they give a command which
transfers that amount of money to the recipient (minus the overhead).
Recipients can pull the money out through their paypal accounts in
increments of at least $5.

I suggest that providing an open-source pinax compatible micro-
payments/tipping mechanism could be useful for many applications.

(2) The second use case is someone spending between $5 and $20 per
month to use a website. I presume payment is through paypal.
I imagine that such a use case is relatively straightforward. Would it
be just a few screens?

Let me know your thoughts on providing an open-source tipping
mechanism with micropayments for pinax.

Jonathan

On May 20, 10:56 am, bobhaugen <bob.hau...@gmail.com> wrote:
> On May 20, 8:32 am, Social Network in DJango <readb...@gmail.com>
> wrote:
>
> > p.s. Bob, would you consider offering your accounting software as a
> > project in pinax?  I will need to be tracking micro payments.
>
> My code will end up being open source.
>
> My goal is to develop a full suite of business apps that work with
> Pinax.  But my process is to develop a series of one-off projects for
> particular user groups and then generalize from there.  In other
> words, from the particular to the general.
>
> Or in still other words, the first iterations will be tailored to
> their particular situations and may not work for yours, but you will
> be welcome to mine code.
>
> Two early iterations are the swaps app in current Pinax, andhttp://code.google.com/p/pinax-cohousing/

bobhaugen

unread,
May 21, 2009, 2:40:51 PM5/21/09
to pinax-business
On May 21, 12:13 pm, Social Network in DJango <readb...@gmail.com>
wrote:
> I suggest two concrete use cases (not necessarily what I'm doing).
> (1)
> I suggest that providing an open-source pinax compatible micro-
> payments/tipping mechanism could be useful for many applications.
>
> (2) The second use case is someone spending between $5 and $20 per
> month to use a website. I presume payment is through paypal.
> I imagine that such a use case is relatively straightforward. Would it
> be just a few screens?

I would prefer to work on use cases that *are* what somebody is
doing. Will you have disclosure restrictions that prevent you from
describing your real use cases, or are they not defined yet?

I did swaps as an experiment to learn Pinax, and a couple of sites are
actually using it, but that was just dumb luck. Everything else I am
doing is for concrete social groups, for their particular needs.

I am happy to help with other peoples' real use cases as I have time,
and I will have a couple of my own, that are not quite nailed down
enough yet to post. Neither of my use cases will involve micro-
payments, but one of them might involve recurring subscription
payments. (Don't know yet for sure.)
Reply all
Reply to author
Forward
0 new messages