Wikimedia Meta-Wiki

Fundraising 2011/Main Page

< Fundraising 2011
2011 Fundraiser Noticeboard
Shortcut:
FR2011/N

This noticeboard is for announcements, updates and testing news related to the 2011 Fundraiser from staff at the Wikimedia Foundation and from representatives of the Wikimedia Chapters

You might be looking:

  • to discuss any announcements or news. Click on the Discuss this beneath the piece of interest to be taken to the appropriate noticeboard section.
  • for the Chapters Portal where chapters can - request tests, review test results, chapter fundraising and general information.
  • to help out translating by joining our team of translators to translate the banners, forms, information pages, and more into different languages.
Fundraising 2011
Report
Noticeboard
Discussion pages
Translation
Chapters portal
Tech, stats and resources
FAQ
Live chat: #wikimedia-fundraisingconnect

August 26 Float/No float & Walling edit count test edit

Today we ran a banner and landing page test.

edit

We used the author appeal from Steven Walling to test different edit counts (number of articles and number of edits) in the banner.

Results: No significant difference

Raw statistics
title impressions views donations amount amount50 click_rate don_per_imp amt_per_imp amt50_per_imp
A: "192 articles" 1496901 23509 328 7290.79 6740.79 0.016900 0.000210 0.004700 0.004336
B: "40,000 edits" 1498921 20335 338 7989.00 6614.00 0.014518 0.000217 0.005132 0.004251
Full report
File:C FloatTest 0826 US banner.pdf

Landing Page Test edit

We tested a new landing page that had the right side of the donation form float in place when you scroll down so you can see the radio buttons even when you're at the bottom of the appeal. Our old page has the radio buttons not show when you scroll all the way down the page to the end of the appeal.

Results: No significant difference.

Raw statistics
title views donations total_clicks amount amount50 avg_donation avg_donation50 don_per_view amt_per_view amt50_per_view
"Float" 21966 326 326 6959.00 6359.00 21.60 19.61 0.014480 0.310204 0.282833
"No Float" 21877 340 340 8320.79 6995.79 23.38 19.89 0.015160 0.370163 0.311622
Full report
File:C FloatTest 0826 US LP.pdf

So no big breaking news findings to report today

Fundraising Drills edit

Posted @ 16:53, 28 August 2011 (UTC)

We have a series of tests scheduled to "go live" today starting at 17:30 UTC.

We will be creating campaigns, testing banners, landing pages, donate now template, asking strings and currencies for: Brazil, Belgium, Columbia, China, Japan, Turkey, Singapore, New Zealand, Poland and Canada. The goal of this drill is not to test any of the creative content or templates. It is just a stress test for the daily tasks and problems we may have in the FR and help establish a more productive and sustainable production process.

The schedule is as follows:

  • 17:30 - 18:30 UTC : Brazil, Belgium, Columbia, China
  • 18:00 - 19:00 UTC : Japan, Turkey
  • 20:30 - 21:30 UTC : Belgium, Japan
  • 20:45 - 21:45 UTC : Singapore, New Zealand
  • 21:00 - 22:00 UTC : Poland, Canada

More details after the test.

Cbarr (WMF)
Production Coodinator

The stress test went well. We had somethings we did right: everyone knew their responsibilities, all the tests we wanted to do ran successfully, no pages with major bugs. Of course there were some problems as well that we discovered a need: to increase the structuring of our communication, for more documentation, to rethink how these pages are created. All and all these tests did exactly what they were designed to do.
Cbarr (WMF) - Production Coodinator 16:19, 25 August 2011 (UTC)[reply]

Discuss this

August 5th, Steven Walling Test edit

Great news! We’ve found our first volunteer appeal to perform on a similar level with our founder appeal. The fundraising team has been testing 2011 fundraising messages for about two months now. We’ve found successful programmer and staff appeals but this is the first editor appeal we have seen perform anywhere near the other appeals.

We tested new banner text with an appeal from long time editor, Steven Walling. Up until now, our best volunteer banner text had been, “Please read: a personal appeal from Wikipedia author Steven Walling.” In this test we tried, “Please read: a personal appeal from an author of 191 Wikipedia articles.” Steven’s volunteer appeal included a new comparison with the other top websites’ number of servers and employees where he said, "Wikipedia, the #5 site on the web, serves 411 million people per month -- with 370 servers and 73 employees."

 
Website Comparison

Key Findings

  • The new banner had about a 20% higher click rate than the Jimmy banner and overall Steven's appeal performed pretty close to Jimmy’s donation level.
  • The clear and direct appeal that emphasizes Wikipedia's utility and efficiency was effective in Steven's editor appeal.
 
Author of 191 Articles
File:Jimmyvssteven191.png
Jimmy Wales Banner

Next Steps

The results from this test are really encouraging. The Storytellers have just returned from Wikimania. While at Wikimedia’s annual conference, we interviewed over 40 editors from around the world with different Wikipedia experiences and stories. Right now we’re in the process of sorting through this mountain of material to start writing fundraising appeals to test. The successful test with Steven’s author appeal is really motivating to find more successful appeals from different editors. We’ll find plenty of new things that work and other things that won’t work, but this year we want the fundraiser to rely as much as possible on voices of many different members of our community.

You didn’t have to attend Wikimania to get involved. We will be interviewing people from all different areas of the world over the next several months. If you're interested in sharing your story and explaining in your own words why donating is important, please send a message to wikistory wikimedia.org.

Raw statistics
title impressions views donations amount amount50 avg_donation avg_donation50 click_rate don_per_imp amt_per_imp amt50_per_imp don_per_view amt_per_view amt50_per_view
Jimmy 1170738 15890 327 6238.54 5338.54 n/a n/a 0.013777 0.000274 0.005221 0.004477 0.020400 0.389744 0.333350
Walling (Plain) 1171567 15577 249 6441.00 5041.00 n/a n/a 0.013478 0.000207 0.005345 0.004191 0.015740 0.408168 0.318220
Walling (191 articles) 1171721 19286 262 5863.00 5188.00 n/a n/a 0.016704 0.000218 0.004874 0.004310 0.013430 0.300300 0.265530
Full report
File:C JMvSW2 0805 US.pdf

July 15, 2011 Ryan Kaldari Test edit

 
Kaldari Banner Images

At this stage in our testing, we’ve found a few successful staff appeals and we’re now looking for effective editor appeals.

Ryan Kaldari is a developer at the Wikimedia Foundation. Before working at the Foundation, he was a volunteer developer and editor for several years. We interviewed Ryan about his volunteering experience and wrote a few different versions of his appeal to test. One version used almost all of Kaldari’s own personal statements, word for word. Another version still used Kaldari’s statements but was slightly more polished and formal and started with the first line, "Wikipedia is the most amazing project that has been created in our generation." The third version was also more edited and had the first line, "Wikipedia makes me feel like I live in the future."' We also tested a variety of banners with different photos of Kaldari and a couple different colored backgrounds.

Results

  • The appeal written using almost all of Kaldari’s own words performed the best in terms of both number of donations and amount raised (between 45-55% increase over the other two versions).
  • The banner with the plain white background had a higher click rate than the banners with the grey and red backgrounds.
  • The banner with Kaldari’s photo cropped in on his face and shoulders had a higher click rate than the banners zoomed out including his arms and torso.

Next Steps

We want to use this practice of interviewing and using personal statements from those interviews to write fundraising appeals. We’re looking for volunteers who are passionate about the work they do and can talk about why people should donate to the Wikimedia Foundation. If you’re interested, please let us know! You can email us at wikistory@wikimedia.org.

Thanks. Mhernandez 06:57, 20 July 2011 (UTC)[reply]

Raw Statistics edit

US test - 6 Banners
bannerlp impressions views donations amount amount50 click_rate don_per_imp amt_per_imp amt50_per_imp
"arms crossed smiling" 636956 8204 34 648.00 598.00 0.012948 0.000062 0.001169 0.001095
"red background" 635692 8060 32 651.33 601.33 0.012801 0.000057 0.001058 0.000984
"smiling shoulders, no arms" 634199 8825 30 800.00 650.00 0.014201 0.000057 0.001501 0.001279
"grey arms crossed" 636944 7882 40 660.00 610.00 0.012379 0.000069 0.001912 0.001370
"hands on hips" 637247 7790 44 1330.00 980.00 0.012419 0.000089 0.002772 0.002123
"white arms crossed" 635336 8411 42 803.33 753.33 0.013225 0.000074 0.001410 0.001336
US test - 2 Landing Pages
lp views donations total_clicks amount50 don_per_view amt_per_view amt50_per_view
"Future" 25089 96 96 1849.33 0.004347 0.095676 0.084841
"Kaldari's exact words" 24080 126 126 2343.33 0.006172 0.162119 0.129318
EN test - 2 Landing Pages
lp views donations total_clicks amount50 don_per_view amt_per_view amt50_per_view
"Amazing project in our generation" 8596 8 8 203.76 0.001012 0.026391 0.026112
"Kaldari's exact words" 8392 17 17 324.27 0.001997 0.037383 0.037240

July 8, 2011 Steven Walling Appeal Test edit

In our latest test, we put a new appeal from Steven Walling up against Jimmy's founder appeal. Steven is currently on staff at the Wikimedia Foundation but he is also a long time contributor. We wanted to isolate the effect of his title, so we ran his same appeal text from two different banners, one calling him a “staffer” and the other calling him an “author.”

 

Results

  • Steven’s “staffer” appeal performed at nearly the same level as Jimmy’s appeal.
  • Steven’s “author” appeal brought in about 20% as many donations as Jimmy’s appeal.
  • Steven's "author" appeal had lower click rate and donation rate than the "staffer" appeal.
  • The appeal text was the same, the only difference being Steven’s title.


This test is encouraging because we've found another Wikipedian's story to tell that performs similar to Jimmy's appeal. We still need to do some work to test new variations of his title.

Next steps

  • We’re coming up with new titles to test for editor appeals. During the 2010 campaign, we tested Wikipedia “editor,” “author,” “contributor,” and “volunteer,” with “author” coming out ahead.
  • We're in the process of interviewing editors to gather a variety of personal Wikipedia stories that can be tested as fundraising appeals.

Reports

Retrieved from "https://meta.wikimedia.org/w/index.php?title=Fundraising_2011/Main_Page&oldid=25879353"