Jonathan Leybovich wrote:
> I am confused, though, as to why PayPal is not a
> viable option. Every payment system imposes some sort
> of per-transaction fee, and the 3% transaction amount
> charge reflects not another part of PayPal's cut, but
> simply the fact that the merchant receives money from
> the card holder's bank immediately while the card
> holder is given a 20-30 day grace period before
> interest accrues on their new balance. Thus switching
> from PayPal may not necessarily save much on costs.
I don't think that
anyone is advocating dropping PayPal or any other
existing payment option, just adding direct donation by credit card as
another option.
At the moment, the options available are PayPal or Moneybookers, either
of which is quick and easy if you already have an account.
However, if they don't have an account with either, a potential donor
will have to:
* go to one of them to start an account
* find out how they work
* decide to trust them with your credit card details
* set up a new account, including filling in captchas, E-mail addresses,
other personal details, inventing __and remembering__ yet another strong
password, reading and agreeing to their contract terms and conditions
* entering your credit card details
* then going through the process of donating to Wikipedia with their
newly set up account
Compare this with a direct credit card donation, where a donor will need to
* decide to trust the Wikipedia Foundation with their credit card details
* entering them and the amount of their donation into the donation form,
and then clicking "OK"
Since donations are mostly an "impulse buy", I would imagine that a
great many potential donors just give up at the first stage of setting
up a new money transfer account.
Thus, direct donations should, if I'm right, significantly increase the
rate of donations by allowing people who simply would not have donated
at all to give money without going to undue effort.
If the CC processor's rates are _significantly_ higher than the
alternative, we can simply add a notice suggesting that people who
already have an account with these prefer using them to a direct
donation. If they are not significantly higher, the message can simply
be left out; anything which requires effort or thought from potential
donors is simply a barrier to donation, and will cause some small
fraction of potential donors not to bother. In any case, I would imagine
that people with these accounts would prefer to use them anyway, since
they require entering fewer details.
At the moment, I would imagine that even a gateway processing
arrangement would be quite effective, since it is intended only to catch
people who otherwise would not donate at all. In the longer term, if the
uptake from CC payments is significant enough, and if a cost/benefit
analysis justifies it, moving to a direct processing arrangement would
be worth considering in the long run.
Here's a specific worked example:
* Merchant Express (
www.merchantexpress.com) quote a rate of 2.33% plus
25 cents, with a monthly subscription fee of $9.95
* assuming a typical donation of $10, that means they take an effective 4.8%
My best guess is that offering CC processing will increase the level of
donations by 15 - 20%.
if only a conservatively estimated 10% more people donate because of the
CC option, increasing a base of $250,000 to $275,000, the net gain per
three-monthly fundraiser would be
95.2% * $25,000 - ($9.95*3) = $23,770
or about ten servers.
Even if CC payment cannibalizes all of the cheaper payment methods, but
still increases donations by 10%, it would still result in a significant
net gain per fundraiser:
95.2% * $275000 - 97% * $250000 = $19,300.
and in the utterly worst case, where CC payment cannibalizes all cheaper
payment methods, and there is no gain at all in donations, the CC
payment option can simply be removed from the donations page at the end
of day one, limiting the worst-case risk for this experiment to (given a
generous estimated $20,000 of donations made on day one)
1.8% * $20,000 + 12 * $9.95 = $479.
...and I'm sure that with an effective monthly average expected take of
$83,333 (on the basis of four $250,000 fundraisers a year) the Wikimedia
Foundation could find/negotiate far better rates than these published
entry-level rates (which were simply the first ones I found via the top
Google hit for "credit card processing").
-- Neil