Talk:Fundraising 2009/Alternative banners

From Meta, a Wikimedia project coordination wiki

Updates on launch plan[edit]

I've been consulting with Rand, Sue, Jay, and the rest of the WMF team on the kick-off of the fundraiser, and wanted to give a quick update on where we are.

First, we're currently debugging the banners to make sure that we eliminate any major browser issues. The issues with IE we experienced at launch are inexcusable, and we'll ensure that those are sorted out.

Second, with regard to the messaging, a brief summary of the plan I've suggested going forward, and some interim comments on the messages that have been suggested on this page.

  • We're not going to go live with anything all-caps. Too many people have specifically pointed this out as annoying and visually unpleasant, and unless we see a major drop in donations as a result of getting rid of it, we're not going to go there again.
  • I've suggested a rotation of several banners to kick things off: 'Wikipedia Forever' (33%), 'For your great, great, great grandson' (16%), 'For your great, great, great granddaughter' (16%) and 'Free Knowledge Forever. Ad-Free Forever. Wikipedia Forever.' (33%). This is only the starting selection for the first phase.
  • We've killed some banners from the early list; specifically, 'This is everything we know' will not run, and 'Look at what you've done' will not run either, at least not without making it a lot more understandable what it means. We'll definitely run the quotes, and we'll probably tweak some of the others. Please keep in mind that changes are tricky to implement and need to be made carefully, as we also need to translate them into many languages.
  • With all due respect, we will not use banners that use witty Wikipedia inside humor that makes us nerds giggle. Disambiguation jokes, stub jokes, and other similar inside jokes, are not suited as fundraising messages for the general Wikipedia-using public. We will also not use 'Please pay for servers' messages: these have a lot of inside appeal (it's the kind of spending the community cares most about), but we want to send a larger message about the mission and the purpose of Wikipedia.
  • I like some of the other suggestions here and I'm glad that some of our original designs are being picked up and played with. We'll continue to review the suggestions here.

The single biggest takeaway from last year's fundraiser is that the large-letter personal appeal from Jimmy performed 10 times better than any other message we've run. We're focusing on messages that have a similar appeal to both gut and brains, and the big picture vision of Wikipedia that Jimmy articulated. We also have an appeal letter ready, and will use it later in the fundraiser.

Our overarching theme this year continues to be, essentially, that we need to ensure that Wikipedia will continue to be around for the long run, that it needs to be protected, that it needs support to be there when you need it. This is also based on our thinking about long term organizational sustainability, and we hope that some of our messaging can potentially be re-connected to in the event that we run an endowment campaign next year.--Eloquence 19:09, 11 November 2009 (UTC)[reply]

Will you be fixing the logo issue? Several years running now this has been an issue. Non-Wikipedia projects should never have a Wikipedia logo plastered across every page.  — Mike.lifeguard | @en.wb 19:26, 11 November 2009 (UTC)[reply]
The easiest way to solve that issue is to not include a logo at all. It's always available in the top-left of the page. Having it twice on every screen is senseless. --MZMcBride 19:28, 11 November 2009 (UTC)[reply]
The plan is to either fix it (use project logo or Wikimedia logo), or to not run the banner on non-WP projects for now.
MZ, I don't agree that having a logo in the banner is senseless. It's useful to have some piece of visual identification in the banners, especially in the launch phase, to ensure instant visual identification of the banner as being part of the site. Last year we used a puzzle piece for the same purpose.--Eloquence 19:35, 11 November 2009 (UTC)[reply]
  • With all due respect, we will not use banners that use witty Wikipedia inside humor that makes us nerds giggle. Disambiguation jokes, stub jokes, and other similar inside jokes, are not suited as fundraising messages for the general Wikipedia-using public. We will also not use 'Please pay for servers' messages: these have a lot of inside appeal (it's the kind of spending the community cares most about), but we want to send a larger message about the mission and the purpose of Wikipedia.
Sites like Twitter and Facebook often use cavalier and coy marketing strategies. Not sure why at least some of the banners can't be a bit clever. It certainly can't hurt to test them out and compare performance. I agree with some of the comments on the subject-space page about terms like "stub": Wikipedia has entered popular culture in a variety of ways, let's use that to our advantage. --MZMcBride 19:32, 11 November 2009 (UTC)[reply]
I agree - let's be creative, and give at least some food for the geekosphere. Having said that, we need other clear messages that appeal more broadly - some suggestions here (eg Mav's) are already improvements. Glad to see the two slogans I was most fearing being axed. Cormaggio @ 10:11, 12 November 2009 (UTC)[reply]
So basically you're going to reintroduce "Wikipedia Forever" despite massive opposition to it, and pretend that the opposition was based solelly on it being in caps? I'm sorry but I don't think thats gona fly. The other messages also don't seem to have much support from what I was able to gather. I'm undecided on whether we should use "inside jokes" as many of them are mainstream now, but the two most voted on this page don´t fit that description. You say you like them, but thats not concrete. People were told to come here and voice their opinions, we did. Are we going to be heard or not? and more importantly are you going to use our suggestions? Acer 19:36, 11 November 2009 (UTC)[reply]

Echoing what I said on the main page: given the traffic volumes, there should be a big element of throwing mud at the wall and seeing what sticks. Serve some decent ideas to X visitors, then ditch or try variations. The returns from finding some really good ideas are high enough that it's worth testing quite a long tail of possibly-good ideas. I'm slightly doubtful about references to "stubs", for instance, but it might just appeal to a certain audience of drive-by editors, who knows? Suck it and see. Depends a bit how hard it is to try lots of variations, but if it isn't feasible now, it should really be on the agenda for next year. Having such a massive user base to test messages on easily is an incredible resource. Rd232 19:39, 11 November 2009 (UTC)[reply]

Also, can I ask explicitly: what exactly is the 2009 funding drive doing to secure the long-term ("forever") future of Wikipedia? It sounds a bizarre promise to make, like a toothbrush giving you eternal life. But if there is a long-term plan being funded, shouldn't the landing page at least try to explain it? Right now it's "Wikipedia forever" / Click / "Huh? What happened to forever?" Why did you use a "forever" hook and then not even try to justify it? Rd232 19:44, 11 November 2009 (UTC)[reply]

As far as I know there's nothing particularly different about the actual fundraising or spending targets from previous years; as before the targets are for covering next year's budget and maybe a little extra. --brion 19:52, 11 November 2009 (UTC)[reply]
So "forever" is just marketing? There's no special long-term planning substance? That doesn't really matter as long as you've got something to point to try and justify the hook. But if you don't even try it makes the claim really weird (quite separate from the linguistic criticisms made, that "Wikipedia forever" sounds tweeny). Rd232 20:10, 11 November 2009 (UTC)[reply]
I've asked Rand to provide some details on this; all I can do myself is point at the FAQ which indicates we're targeting about 3/4 of this fiscal year's budgeted expenses for the public donation campaign. --brion 20:21, 11 November 2009 (UTC)[reply]
Thanks. Perhaps you could address the first of my two points as well - how easy is it to test lots of messages and variations on messages? I've used Google Ads in the past, and it does help. And it's a lot more effective if you've got Wikipedia-scale traffic... Rd232 21:52, 11 November 2009 (UTC)[reply]
The theory is certainly to test a bunch and see what response they get. --brion 21:58, 11 November 2009 (UTC)[reply]
The infrastructure, as I understand it, allows one to test an arbitrarily large number of variations. Though as the testing pool increases so does the complexity of handling translations, unless you forgo that step entirely, and the number of checks to make sure each new version works in every browser, etc. To get a rough handle on relative performance one probably needs about the equivalent of 4-6 hours of full traffic (or equivalently 15-25% traffic for a full day), so one could hypothetically test several variations per day and evolve the messaging rapidly in response to performance. Historically we've generally changed every few weeks of a campaign. Dragons flight 22:25, 11 November 2009 (UTC)[reply]
I suggest eliminating all of that excessive vertical whitespace that showed up above and below the banner's text. That looked incredibly amateurish. 22:40, 11 November 2009 (UTC)


We will also not use 'Please pay for servers' messages: these have a lot of inside appeal (it's the kind of spending the community cares most about), but we want to send a larger message about the mission and the purpose of Wikipedia.

So, uh, maybe I missed this, but what exactly are we asking for money for, other than servers and associated running costs (eg staff)? Furthermore, this is supposed to be a fundraising scheme, not some form of PR. We're after money, not ideological converts. Modest Genius 23:08, 11 November 2009 (UTC)[reply]

Exactly the point I've made elsewhere. The Foundation is asking for money; after paying for servers & bandwidth, what is it going to do with the money??? Tell those folks you want to give exactly what will be done with it -- otherwise, they'll just think it's all going into Jimmy Wales' pockets (even if it isn't) & decide to keep it in their own. -- Llywrch 07:14, 12 November 2009 (UTC)[reply]
Surely the point Erik was making is that we will be using the money for servers (amongst other things, such as salaries), but that appealing to the general public on these grounds is less effective than on a wider message of 'free knowledge'? Cormaggio @ 09:59, 12 November 2009 (UTC)[reply]
I don't think we have any data on this from last year's banners... As I understand, the only one that seemed to do significantly better than others was the direct personal appeal ("Please read: a personal appeal from Wikipedia Founder Jimmy Wales"), which isn't represented in the current campaign. Nor did any of last year's banners mention what we're asking money for, though some generically indicated that Wikipedia "needs your support" or "is a non-profit", neither of which performed particularly well. --brion 17:18, 12 November 2009 (UTC)[reply]

  • I don't think that any of the current slogans are very suitable as-is. For one, they don't give any indication that it has to do with a fundraiser. Without clicking on the link, a casual reader might think we were bragging about ourselves - or at the least, it would be confusing and make us look somewhat silly. The least we could do is to put a "Donate today" or "Please support" tag below them. That would clarify that this is a fundraiser, and perhaps more people would be encouraged to click the link and donate. Tempodivalse [talk] 23:12, 11 November 2009 (UTC)[reply]
Indeed, one complaint I heard from standard users who are not Wikipedia community members (friends, family etc) is that it was confusing and didn't convey any meaning, or even that they *should* click on the banner. They just thought it was triumphant self-advertising. Orderinchaos 01:21, 12 November 2009 (UTC)[reply]
I agree. Omitting that kind of standard action message from an ad is either really clever or really stupid (depending on whether it works!). It's certainly unusual. Rd232 01:32, 12 November 2009 (UTC)[reply]

End of (work) Day update for 11/11/09 PST[edit]

Lots of progress made today:

  1. Working on the IE6 and related browser glitches. Mostly fixed (some banners will just always look funky in some browsers). Final testing of banners tomorrow morning.
  2. Un-capitalizing all the banners. (Done and done: http://meta.wikimedia.org/w/index.php?title=Special:NoticeTemplate/view&template=2009_Notice1&wpUserLanguage=all). And new banners will not be all caps.
  3. Re-working the display and back code for the credit card payment system. Half done. More adjustments were made. More on the morrow...none of the remaining issues are showstopping.
  4. Another critical issue was resolved regarding our Anonymous/non-Anonymous showing on our Donor Comments page.
  5. The Geo IP testing and implementation. This is working well, but we still expect a slow launch. It's tested, but we need to review under actual load conditions. Want to try it?
Click on this: http://meta.wikimedia.org/wiki/Special:GeoLite?lang=fr&?utm_source=2009_Notice11&utm_medium=sitenotice&utm_campaign=fundraiser2009
If you live in France, you should see this page: http://wikimediafoundation.org/wiki/Global_Support/fr . That's the French Chapter donation page.
If you don't live in France, you should see the regular French donation page: http://wikimediafoundation.org/wiki/Support_Wikipedia/fr

What's happening Thursday?

  1. Last testing of banners on different browsers.
  2. We relaunch on Wikipedia only with the following messages: 'Wikipedia Forever' (33%), 'For your great, great, great grandson' (16%), 'For your great, great, great granddaughter' (16%) and 'Free Knowledge Forever. Ad-Free Forever. Wikipedia Forever.' (33%). (Maybe someone could link to them from the staging page?).
  3. We monitor GEO IP settings and load in France as well as overall performance of our banners.
  4. We work on tracking numbers for our banners and buttons.
  5. We work on the infamous Ariel Extension (keeps users on their project pages when they give). It could go live shortly.
  6. We continue work on project specific banners with project specific logos.

I appreciate all your help and comments. Please feel free to post on this page or here on any issues or questions. Rand Montoya 03:34, 12 November 2009 (UTC)[reply]

thanks for the updates! quick question: have the "For your great, great, great" banners already been pulled? I can't get them to come up. -- phoebe 03:49, 14 November 2009 (UTC)[reply]
Yes, i think so (thankfully!). I can't get them either. Tempodivalse [talk] 04:04, 14 November 2009 (UTC)[reply]

Choosing community-made banners for the live campaign[edit]

We need to begin discussing how community-made banners will be incorporated into the live campaign. There seem to be quite a few banners that have strong support on this page. Does anyone have any thoughts about this? --MZMcBride 05:01, 13 November 2009 (UTC)[reply]

Have a poll & those that make the cut get used. Voting is not always evil. -- Llywrch 05:56, 13 November 2009 (UTC)[reply]
So we first need to gain consensus about how we'll determine consensus on which banners to use? Surely the WMF can use their discretion (if they don't want a simple vote per user, like suggested above). In the interest of saving time. Anrie 10:05, 13 November 2009 (UTC)[reply]
Perhaps that $250k could cover skimming a page, identifying good ideas, and perhaps coming up with a more structured way to get community input. Rd232 10:08, 13 November 2009 (UTC)[reply]
No that would've been the $251k package.
A poll seems like the most straightforward way. But will the foundation listen to us now? It hasn't so far. Tempodivalse [talk] 14:33, 13 November 2009 (UTC)[reply]

Take a poll. It will still show more consensus than the Foundation's been interested in. Fleetflame 02:26, 14 November 2009 (UTC)[reply]

So Fundraising 2009/Alternative banners/Vote then.Geni 13:49, 15 November 2009 (UTC)[reply]

An update on the fundraiser[edit]

Hello all,

I wanted to give you an update on where we are with the fundraiser and our messaging strategy.

As you have noticed, we're now fully live (on Wikipedia only for now) and will stay up until the end of the fundraiser. Please report technical issues here: Fundraising 2009/Launch Feedback. Please continue to post banner suggestions here: Fundraising 2009/Alternative banners.

First, a few comments on the theme "Wikipedia Forever", and why we chose it for this year's campaign. "Wikipedia Forever" is meant to represent both a present and future commitment from the Wikimedia Foundation to preserve, protect, and support Wikipedia for the long run, and to appeal to the public to support us in doing so. Yes, it's an emotional appeal, not a literal message. It's our kick-off message, not our only message. It will be followed by other messages later in the campaign which will re-connect to the same theme.

Our fundraising messages have, historically, always caused significant unhappiness among the editor community. However, we also acknowledge that the early, all-caps, "WIKIPEDIA FOREVER" banner was too in-your-face for many Wikipedians, and that even the current "Wikipedia Forever" theme, as part of a larger set of banners, still raises concerns. Please be assured that we're listening. Again, the "Wikipedia Forever" theme is only the kick-off of the campaign, not the entire campaign. We've already eliminated all-caps from all our banners, and we've pulled up some of our later-stage banners into the first phase. This represents a change in response to the feedback we've received from you.

The guiding principle of our campaign is data. As we're gathering the first full reports about the success of the banners we're currently running, we'll be able to make a better decision, going forward, about which ones to continue to run. This is the primary driver of the campaign.

Will we run community-developed banner ideas? Definitely. Please suggest such banner ideas, and vote or comment on them on the wiki page. If the banners connect to the idea of long-term protection, ensuring the survival of Wikipedia, that would fit into our overall theme.

We will take this feedback into account, and add at least 1-2 of them to our banner rotation. Please understand that testing new banners isn't free to us. They have to be properly implemented, browser-tested, and if they work, translated into many languages, which requires coordination and support from both staff and volunteers. A single banner can amount to many person hours of total work.

Please peruse the following links for data from last year's fundraiser which may inform your thinking about banner design: [1] [2]

Please also understand that, at the end of the day, we'll have to apply our best judgment as to what to run and what not to run, based on data, experience, and conventional wisdom.

I want to address a few common comments on the campaign so far directly:

- "There's no invitation to donate". That's intentional. Our intent with the first round of banners was to build interest and momentum, get people to start thinking and talking about the campaign. We'll have versions with buttons up shortly. We expect that those will generate more clicks and more donations, as our past data has shown that buttons and thermometers increase donations. We're also building a traditional thermometer notice to test. And we have a whole new range of banners coming up, including a personal appeal from Jimmy later in the campaign.

- "What are you actually doing to protect 'Wikipedia Forever'?" Everything WMF does is designed to create a sustainable support infrastructure for the long run. To give specific examples, there are parts of WMF's day-to-day operations that directly contribute to long-term preservation and access, such is the improved availability of full downloadable backups of all Wikimedia content. We've made significant improvements over the last year: regular generation of all dumps except for the full-history en.wp dump, additional off-site backups, public datasets mirrored through Amazon.com. We'll continue to invest substantial resources in these programs. We are supporting various offline content initiatives, which also contributes to the "lots of copies keep stuff safe" principle. We're also planning an additional data-center to protect Wikipedia against disaster and hardware failure (which would have less long term impact, but a high short term impact). And for the long term, our annual budget includes reserve funds. These initially will help to build our short term contingency, and over time will hopefully grow into a veritable endowment fund. The "Wikipedia Forever" theme was partially chosen with an eye to running a future campaign, online or offline, to support such a fund.

- "You guys wasted $250K on a PR company to do this." We have an 11-month relationship with Fenton Communications. This relationship is scheduled to continue until July 2010, and includes a wide range of communications support for all our public messaging activities. Fenton was selected through a request for propsals, due to their comprehensive portfolio working with progressive non-profit organzations. For the purpose of supporting our fundraiser, Fenton is partnering with Sea Change Strategies, a market research and fundraising company.

Fenton and the WMF collaborated on developing the messages for the campaign. Fenton has been reaching out to volunteers, board members, stakeholders, staff, and like-minded organizations to gather facts and opinions. They presented that information to us, and with more feedback from us presented high level ideas that we refined and further developed together. Aside from the direct fundraiser work, Fenton has also been working with the Foundation's communications team to develop a plan that helps spread the messages about the campaign and the Foundation's work through mass media and social media around the world. So the $250K isn't paying for one campaign concept, 'Wikipedia Forever.' The budget covers 11 months of big picture communications and PR support across the board from two different agencies.

We'll continue our collaboration with the Fenton and Sea Change team as we fine-tune the campaign and blend in community feedback. They're experienced in adapting concepts and campaigns to shifting conditions, and we still have months ahead to consider.

- "You're not listening." We are listening, and we've already acted based on some of the feedback received. This includes the feedback in October, which was a reason for us to immediately eliminate two banners from the original set. We've listened now, which is why we've toned down and diversified the banners. We will iterate and respond based also on what the data tells us. Again, this iteration may not happen as quickly as you'd like it to: people here are already working 12-hour days to make this fundraiser a success, and we do what we can to iterate and respond, all the while continually improving the fundraising machinery as a whole.

Other action points for the next week include:

  • improvements to geographic targeting
  • improvements to the credit card form
  • roll-out of a first set of messages in non-WP projects (these will not say 'Wikipedia' nor include the Wikipedia logo)

Rand Montoya, who is running the fundraiser, will post further updates as things develop. Please assume good faith, and let's work together to make this a success.

Thanks and all best,
Eloquence 00:47, 14 November 2009 (UTC)[reply]

Thank you for the commitment to localize both project names and logos. It is disappointing to see that that wasn't done from the get-go since this lesson had supposedly been learned on an annual basis for several years running. But for the future, that is very good news.
Thanks also for pointing out how much work may go into a single banner. I know it is frustrating to users that ideas cannot be implemented immediately, and I'm sure it is frustrating to staff as well - but knowing this makes cooperation much easier.
 — Mike.lifeguard | @en.wb 01:35, 14 November 2009 (UTC)[reply]
When do you expect there to be some data available regarding response for this first batch of messages? Tony Fox 08:05, 14 November 2009 (UTC)[reply]
Re the "what are you actually doing to protect 'forever'" - some of that is reasonable. But it's still a long way from justifying the "forever" theme. And the donation page [3] still needs to try to justify that theme - at least translating the points made above into a few simple words. Rd232 11:06, 14 November 2009 (UTC)[reply]
I'm missing a comment regarding the lack of awareness of cultural differences and the totalitarian feel of the campaign. If the impression it makes outside of the US is so unimportant, why not just run the campaign in the US only / on enwiki only? --Pjacobi 11:47, 14 November 2009 (UTC)[reply]
Thank you for this excellent update. I'm really looking forward to the campaign, especially the data-gathering elements of it; and have been impressed by the responsiveness of everyone involved since the campaign began, considering the amounts of work involved. The localized messages will be well-received! Sj+ translate 01:05, 15 November 2009 (UTC)[reply]
Re "Yes, it's an emotional appeal, not a literal message." - for clarity, I don't think the complaints have been about it being an emotional appeal per se. It seems to me that the complaints are due to it appealing to a different emotion than what was intended. You mention the intent is to represent a commitment "to preserve, protect, and support Wikipedia for the long run". But it instead reads like a proclamation of internal allegiance and unity from some vague outside danger (i.e. en:ingroup). Thus, there are complaints about totalitarian overtones and it being unrepresentative of Wikipedia. AySz88 03:30, 16 November 2009 (UTC)[reply]
That and I'm surprised Eloquence, as a German, doesn't think the slogans in German sound like something Erich Honecker would have said; as an American even I got that feeling. Mike Halterman 13:10, 16 November 2009 (UTC)[reply]
For the purpose of translations, this is why we invited translators to a) tag translations if they feel they aren't working well, b) not necessarily do literal translations. This is indeed what happened in several languages, e.g., the German Wikipedia banner says "Strengthen Wikipedia for the future". I don't think that slogan has totalitarian overtones.--Eloquence 18:48, 16 November 2009 (UTC)[reply]
Super-quick general update: we're working on implementing and testing new banners today, and hope to have new ones running at end of day pacific time.---Eloquence 18:48, 16 November 2009 (UTC)[reply]

Update for Monday 11/16 PST[edit]

Hey All--

Acknowledging that our early fundraising efforts extremely slow (less than 50% from last year) so we are switching up messaging and testing some new messages and approaches.

In addition, we continue to make strides to our significant technical hurdles:

  1. GEO IP: I believe, but could be wrong, that this is working well. I continue to monitor things here. I'm finally going to knock this one off the list.
  2. Credit Card Payment gateway page has been re-skinned (in addition to other improvements) for an even better donor experience.
  3. Tracking: We continue to make progress towards both cleaning up some errant data and providing accurate public reporting. I expect that we will have something presentable on Wednesday at the earliest.
  4. The titles of our primary donation page on WikimediaFoundation.org is no longer visible and annoying.
  5. New Site Notices: After a weekend of data, we have significantly adjusted our messaging to both test and explore how our users respond to different messages.

We are running these on Non-English Wikipedia sites:

We are now in the following rotation on English Wikipedia:

Still on the list for tomorrow (smaller items not included):

  1. Tracking!
  2. Cleaning up our errant data.
  3. Plan next set of banners including a possible community banner. We should have a post up soon @alternate banners for process.
  4. The infamous and dreaded Ariel Extension.
  5. Project Specific Banner
  6. Discussion of solution to Javascript banners and donation page.
  7. Do reasonable followup on some possible fraudulent donations.

As always, I appreciate your questions and comments. Rand Montoya 04:11, 17 November 2009 (UTC)[reply]