Re: [Wikitech-l] Re: Credit card processing for forthcoming

Jonathan Leybovich wrote: 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