Donations are still coming in, but not at a very high rate. We can probably make the $50K goal, but we should still consider how to make our fundraising campaign more effective.

One as of yet unanswered question is how many subscriptions we will get because of this campaign, which may well play a substantial role in supporting us in the long run. However, from cursory checks, the subscription rate appeared to be fairly low.

I have studied online fundraising campaigns a bit. Here are some quick thoughts. Some of these things can be done right now, some require technical changes.

1) Real-time updates. We now do manual updates within regular intervals, which is good, but real-time updates generate TV-like excitement. Kuro5hin.org raised $35K in a few days using a realtime-updated progress- bar shown on every change. (K5 is much, much smaller than we are, but of course there's the added novelty effect when they did it.) This can be combined with..

2) Immediately visible donation comments, increasing the community feel.

3) Progress-meter and PayPal link must be visible on every page, not just a separate page, i.e. like this:

[==================== ] $50K Last 5 comments Wikipedia rocks! Donate [10 ] [^ EUR] [ via PayPal ] [Go] Keep it coming .. ...

(You are using a non-proportional email font, right?) Note that this example includes all the form elements to make a donation with one click, with a reasonable value prefilled into the entry field. This is essential - click-through forms result in hesitation and second thoughts, and then there is the essential laziness of the web surfing process. Also, people like to click on buttons.

This will take up a fair bit of screen estate, which is annoying, but I consider it necessary. We can still support turning it off via CSS, but for the duration of the campaign, we really need to grab people's attention.

For extra funkiness, one could us something like mod_pubsub: http://www.mod-pubsub.org/ to do actual *real-time* real-time updates, i.e. the bar moves and comments come in even as you stay on one page. This would probably be best reserved for a separate fundraising page, though.

4) Make the form/progressbar available to external sites - if we have real-time updates, we should try to make it reasonably easy to include a graphical bar + donation form elements in your blog, on your personal homepage etc. This way we can get the whole blogosphere involved in the fundraising process.

5) Surprises - to keep the campaign exciting, surprises could be unveiled at certain milestones: articles, images, links, anything. Or maybe a "Meet the Wikimedians" series where different Wikimedia users are introduced every $1000. I'm sure we can come up with lots of ideas.

Effectively, the fundraising camapign itself has to be so interesting that people will want to check it at least once daily, if not multiple times. This worked very well for the Dean campaign where they had an ongoing blog that tied directly into the fundraisers.

6) E-mail - the Dean campaign also used a huge list of email addresses for fundraising alerts. This of course has to be strictly opt-in, but could have an additional outreach effect.

I'm reluctant to propose other mechanisms used by the Dean campaign, such as affiliate donation boxes on user pages, because I don't think we should target the Wikimedia contributor community too much.

--- Caveat:

We are in the position where the people who *should* support us often don't know who we are. They may only have a vague idea what Wikipedia is based on reading Wikipedia articles from various mirrors and occasionally from our site - they may consider us equivalentto fact-index.com, thefreedictionary.com etc. They may be just as willing to donate to these sites as to us.

The people who are most targeted by any fundraising campaign are unfortunately our regular contributors, because they generate many pageviews. How to solve this dilemma? Ideas:

* Give signed in users a convenient [hide] link for obnoxious fundraising headers * Raise awareness of the Wikipedia brand by - more strictly enforcing the GFDL - petitioning Google (first privately, then publicly) to give us better treatment, since we are the original source - creating an official Wikimedia affiliate program for mirrors, which would include some logos etc. to make people more aware that articles are from Wikipedia -- in return, give member sites easier ways to update their database

---

Technical notes:

Real-time stuff of course makes caching harder. However, using Edge Side Includes (ESI), we should be able to mark-up only the relevant part of the page as dynamic, and cache the rest.

Existing code: This Drupal module contains some PHP code on PayPal IPN handling that may be useful: http://drupal.org/project/paypal_framework

- - - - - - - -

Now, one might make an argument that unobtrusive ads are preferable over frequent obtrusive donation campaigns. Right now I have no strong opinion either way, but based on the data available to me, I believe that a high- profile fundraising campaign could be over fairly quickly with high returns.

One big problem this year is of course that it's a hotly contested US election, so many Americans have given hundreds of dollars already and don't have much money to spare for things like Wikimedia. We should consider this a good thing, because we need to put our donation model to the test properly, and this is a good opportunity to do so.

Oh, and we of course need to still go for the big money in the form of institutional grants and corporate donations. The collaborative volunteer model may not be good enough here - it's fairly meticulous work where it's often a good idea to have one person on the job. Paying someone a couple grand to do this properly may provide huge returns that make our $50K campaign pale in comparison.

Apologies if I've missed previous discussions on these matters - as ever so often, I'm just throwing ideas out there in the hope of contributing to finding better solutions.

A final suggestion that Anthere should like: I believe that after the fundraising campaign, we should do an international poll among Wikimedians to figure out - why they donated, if they did - why they didn't donate, if they didn't - whether they told anyone else about the campaign etc.

Regards,

Erik