Making Wikipedia profitable

You might have noticed that this is a dotcom site, Wikipedia has to make money. This page is intended to be a forum to knock out ideas to get Wikipedia a profit.

Well, no, Wikipedia doesn't have to make money--except insofar as it wants to support its server and employees who work on developing various aspects of the site. Moreover, because this is an open content/"free" website, anyone can take the contents of Wikipedia and proceed to (try to) make money from it. So...

---

Tried, Tested and Tired Methods

• banner adverts - generally awful, not proving to work very well at all. Click through rates are low and banners are generally considered a nuisance at best.

Unfortunately, you happen to be wrong about the profitability of banner ads.  :-/

• subscription - free or otherwise, a bad idea, certainly worse than banner ads.

I totally agree.

Exciting new 21st Century Methods

• branded merchandising - t-shirts are always popular amongst geeks, print up some of the most favourite and oddball pages and sell hundreds.

Put me on the list for a t-shirt! --ChuckSmith

• micro-payment system - add a small donation button to the top and tail of each page.
• Wikipedia cd/dvd - wait a while, bring across the majority of the well authored and editied pages to cd/dvd and add in loads and loads of multimedia content; videos, music, pictures. Package it up, sell it for a reasonable price and watch the money roll in! This is probably one of the best ways for Wikipedia to make money in the long run, there are still billions of people without highspeed Internet access (or any at all) and the added advantage of multimedia would even make it a draw for those who can get to the website.
• permit /Advertisements or /Ads subpages
• adapt Google's AdWords concept
• any more?

---

How about an Experts for Hire page?

So-called expert sites like guru.com and exp.com get a cut from the micropayments (think paypal.com) charged by their "volunteer" subject-matter experts.

See N.Y. Times article by Lisa Guernsey, February 3, 2000 Suddenly, Everybody's an Expert on Everything: Sites Turn Questions and Answers Into a Free-for-All, but Sometimes the Facts Get Trampled for an unvarnished look at this idea.

---

Easy method: Contribute useful articles so that everyone can profit from the concise knowledge contained therein.

...Except this doesn't actually translate into real money that can be used to keep ther server running, by bigger hard disks etc. sadly most people look no further than placing those awful banners on every page and then wonder how to coerce the readers into clicking said accursed ads.

Making Wikipedia profitable/Talk

---

Jimbo here...

One interesting angle for Wikipedia is that each page is definitively keyword-based. That is, the title of the page serves as a very clear and simple keyword. This enables us to hook up with some keyword-based advertisers. People like Go'To.com and Directhit.Com. We use these advertisers with great success at Bomis.

The advantages, I would hope, would be that the advertising would actually be relevant to the page in question. Right now, keyword based advertising is still in its infancy, but in the future, I can imagine that a page on World War II might magically have an ad which corresponds to that topic -- perhaps an ad for a video series on that topic, or for books on that topic.

One of the most important facts about Wikipedia is that it is really a very low cost way to generate content. What do I give up in exchange for the free content everyone is putting on here? Well, the most important thing is the *license*, which guarantees that your work will never be proprietary. If people don't like the way we put ads up here, if we try to overcommercialize it, then someone can very easily just take the content and put it up for free elsewhere.

This limits the upside of profitability, obviously. But due to the overall low cost of running this, that's o.k.

---

A semi-brilliant way to generate keywords automatically: for each page, the page's keywords are defined by the set of the words in the titles of other pages that link to the page and of other pages to which the page links. --LMS

---

I think I'd probably stop contributing and telling my friends to contribute if we started running ads here. I see an ad as a withdrawal from the social bank account.

RANT

Let me explain with a better example: door-to-door salesmen. Once upon a time people had strong notions of hospitality when visitors came calling. When someone came to see you in person, you invited them in, offered them your hospitality, and listened politely to them.

Some brilliant marketer decided that this was a great opportunity for him to exploit - he could act as if he were calling on people as friends, and give them a hard-sell for his pet product. This gave him access to people who would not normally take time to hear his pitch, and in that population was a percentage who were pushed over the edge by this and bought his product. That is, door-to-door sales worked.

Now, when was the last time you invited a salesman in and listened to what he had to say as if he were a "normal" guest? Many people, myself included, have been taken once or twice on a product that wasn't really worth it, and now we're automatically suspicious of everything visitors have to say.

I haven't run the numbers, but I suspect that if I did I would see a decline in the effectiveness of door-to-door over time, as people wised up. Somebody made a lot of money before that happened, and that's fine for him, right? Well, I don't think so, because the by-product has been a less trusting and hospitable society overall.

I felt this personally when I did some church ministry work. I had nothing to sell, and my goal was to share philosophies which I found effective with people who I considered to be spiritual brothers and sisters. No hard sell, no ulterior motives, just that. Yet virtually everyone I spoke with had that look which said "you're a salesman, and a nuisance by default". Notably, immigrant populations who hadn't grown up in a door-to-door scam culture generally treated us much more like normal people, and we got along much better.

So the way I look at it, we all have ended up paying the tab for the sales that were made which shouldn't have otherwise happened. Remarkably enough, I got two phone calls while writing this response. The caller ID reported "Unknown name", so each time I thought it might be my family calling, since they have caller ID blocked. Each asked for a roommate by name, and I had to ask whether they were selling something. Both were, as I had guessed by their I'm-reading-a-script voices. So I figured "What the heck?" and asked them if they'd like to hear what I had written. At that point they began speaking much more like normal human beings and asked for the address of the wikipedia. (Then I asked them nicely to put me on the do-not-call list).

Anyway, that's my general outlook on that kind of marketing. My university drops magazine ads (whose content is quite opposed to the principles of the university) into my bookstore purchases. They get money for that, and I have less respect for them in return. Likewise, if I read an article on World War II in the pedia and see an ad for a set of books which turn out not to be that great, then I'm annoyed the next time I visit the wiki. Perhaps a trivial amount of tarnish to the wiki's reputation, but in exchange for a likewise trivial amount of money.

/RANT

Anyway, if we did go to ads I'd be very tempted to fork the project to my own wiki. (I guess that sounds like a threat, which it really isn't intended to be. I'd just hate to see the open content everyone's written be surrounded with a bunch of advertising slogans from companies that are paying for banners precisely because they don't understand the wiki nature.) -J

I think a better analogy would be of someone renting a large hall for a community event, in exchange for advertising during the event. If you can provide a better meeting place, please do it. Reliable hosting for CPU-intensive services is not cheap. (I went through two cheap hosting companies that couldn't handle even the minimal wiki traffic of usemod.com.) Personally, I am greatly impressed by the openness of the Wikipedia. Even most "free" sites don't allow users to copy the full content for another site. --CliffordAdams (author of UseModWiki)

To me, that still has the same kinds of problems, depending how it's done. It's selling part of the pedia's reputation to a random company. Especially when it's the kind of advertising that's like "Now Johnny will play 'America, the Beautiful' for us on this nice Kawai(tm) piano. BTW, he keeps in top playing condition by eating Wheaties(tm), the breakfast of musical champions!" because Kawai and Wheaties were the highest bidders. Something more like "this site sponsored by Andover.net" down at the bottom of the page seems, OTOH, to my twisted fancies much more acceptable - perhaps because it's more honest. They help us out, and we're grateful. But they're not demanding our attention or trying to manipulate our sensibilities.

I (Jimbo Wales) totally agree that for the Wikipedia, in-content advertising would be disastrous. Also, it would be disastrous and a grave moral breech on my part if I did anything to violate the trust that the authors have placed in me. The content needs to remain in the hands of the community, of which I am but one member. If I added an advertisement to the content of the site, anyone could (and should!) delete it!

But consider a page about Ernest Hemingway. Why not link to places where interested readers can buy his books? And why not have those be affiliate program links. Even here, I would say that it only makes sense for me to try to avoid merchants who have offended the free content and free software community (Amazon, with their one-click patent nonsense, for example).

As to the expenses, the major expense of this site is, and will continue to be, Larry's salary. I could fire Larry, and the site would still grow, more or less. But I don't think it would be as good without his excellent full-time coaching of us all.

Larry immediately breaks into a cold sweat.

How much does it cost per year to run the Wikipedia? More than USD 100,000?

How much does the pedia use in the way of system resources, anyway? My friend at zayda.com who hosts my wikis originally turned me on to this site, and might be interested in having the pedia there.

---

Right now Wikipedia doesn't use a significant amount of server resources (though it does use a significant amount of paid human resources :-) )--and it still couldn't generate very much ad revenue, either. But eventually, it's quite possible that it will require several dedicated servers as well as a staff to run them and to keep the site running smoothly. That, of course, will cost money. I suspect Jimbo might, to help appease those who are morally opposed to ads (for whatever reason), make it possible to make a cookie setting so that ads are turned off or automatically minimized, or something like that. That's just an idea. Anyway, no, we aren't interested in having Wikipedia hosted anywhere else. --LMS

---

I think something worth at least knowing is where the money is intended to go. The above lists upgrading and maintaing wikipedia as sample destinations, but make money presumably means a lot of it is intended as a reward for creating the site.

The money will go first and foremost to pay Larry's salary. Anything above and beyond that will go to support my extravagant lifestyle. If you want, you can donate money to this cause, as well. The Bomis Tipjar --Jimbo Wales (yes, this is humor)

---

I just noticed that when someone writes ISBN something, like ISBN 0691004021, links to barnes & noble, amazon and pricescan appear, automatically. Is that part of "making wikipedia profitable"?

No, it isn't. That's always been that way... Clifford put it in there. I could change the software to make those links into affiliate program links, ad we'd make a little money off of them. Maybe I should do that? But, I'm very uncomfortable with any mixing of _content_ and _revenue_. I don't want people to feel that there's a conflict of interest in the content part of the site. Ads should be clearly marked and separate from the content, and should not impact the content at all. That's crucial for the integrity of what we are all working towards here.

I doubt commissions off of book sales would be offensive to that many people. It has become an established custom on the net. Any link to a book that does not meet standards of topicality would be deleted by random editors. Links that add value to the content because of their topicality would persist. Advertising is information and has the potential to be very useful - in a wealthy society with myriads of options we would be lost without mechanisms to make us aware of the options that exist. The reason advertising is so annoying is because most of the time the information we receive is not wanted, because it is broadcast rather than targeted. Well-targeted, topical ads, of which book links are a very good example, have legitimate information value. I know from experience how little money we could hope to make from these links, so making money is not a good reason to add them. Adding value for site users would be a good reason to add them. - TS

I agree, and these would be totally inoffensive if the words "Amazon" and "Pricescan" did not appear in the article itself. But if the ISBN simply appeared as a link, linking to a local page describing an affiliate program where one could buy the book, that would be useful and totally appropriate. In fact, I'd rather have the affiliate program link than have the advertising appear in the text. --LDC

I generally agree with the ideas above. The ISBN links go to booksellers because they are the best free source of information I could find for the links. Originally I intended to allow the user to select where the ISBN link points to, but its a low-priority feature request. One problem to consider with affiliate programs is that they are often exclusive--not what one would want in an encyclopedia.

Perhaps a short-term simple fix would be to change the names of the additional links to "alternate" (for Amazon) and "search" (for pricewatch). In fact, I just now made that change for the next UseModWiki release (0.92)--I was always uneasy about the "Amazon" text since I'm personally boycotting them. --CliffordAdams

Just in case someone doesn't know what boycott is Clifford talking about, see http://www.gnu.org/philosophy/amazon.html

For ISBN links, perhaps instead of putting direct links to Amazon et al in the content page, it could go to a new page. The page could list informational and purchasing links and indicate whether they were affiliate programs, and explain the purpose of the revenue, etc. --Alan Millar

---

Comment written by LA2 on May 10, 2001: I just wrote a page on the dotcom death. The only sort of financing that I believe in for a site like this is a fat, one-time donation from someone who has too much money. Don't tax the users with subscription fees or advertisments. These plans are doomed. I didn't know some of the contributors here were paid. I'm certainly not. I just spent a few evenings here writing articles on everything conceivable and I was surprised that I got to write articles on things like San Francisco (not exactly my field of expertise). Then I found http://www.nupedia.com/ which has published 17 articles in 10 months (not really a threat to Encyclopaedia Britannica), so Wikipedia is the big and dominating project, and in just a few evenings I have been able to write more articles than most people here (including LA2, IP 130.236.221.xxx on May 3, and IP 213.112.220.xxx on May 10). That is sort of embarassing, and shows how thin this Internet is, still after 30 years.

---

No one is paid to contribute here. Larry is paid to edit and organize; he also happens to have contributed, just as you and I have. I also doubt very seriously that you have written more than most people here (though you may well hold a record for the number of contibutions to different subjects in a small time). The server does not retain statistics on who writes how much, which befits this sites general lack of concern for the concept of "authorship", which is what I like most about it. Contributors will come and go; their text will continue to accumulate. Yes, it will likely take years before we have anything vaguely competitive with EB, but we will get there, and pass them up, and it will have value that may provide some innocuous way to support itself without sacrificing utility or freedom. --LDC.

---

Here's my thoughts, coming from someone involved in other open source endeavors. First, I think we could generally say that if Wikipedia were to start making a profit, some authors would feel cheated, or that the site had "sold out" and was no longer sufficiently community-controlled. Thus I would suggest approaching it strictly from a non-profit standpoint. Period.

Becoming a non-profit brings with it certain other advantages. People are accustomed to the notion of needing to donate money to non-profits, so donation-based schemes are complementary. Similarly with grants from other companies, to which I will comment further on since no one else has yet. If the entity were to not meet costs, then it will be viewed as a failure of the community to support the site (or of management to properly manage funds) rather than an issue with the business model. And the tax situation becomes better - both for the company and for the donors. And it clarifies and puts to rest concerns over what the money is being used for. There is a vast number of non-profits out there, one of which could surely provide a decent model to base Wikipedia on.

Of course, non-profit does not mean non-cost, and some means of acquiring income is necessary. Donation is actually a viable approach; it has supported GNU for years. But IMHO, grants are probably a more fruitful source of reliable funding in the near term. As an educational resource, Wikipedia will be of interest to schools and libraries, some of whom could potentially be tempted to provide small grants in support of it (the idea of not *having* to pay, but *needing* to should appeal to the museum/church mindset). As a free public reference system it could be of interest to various governments, medical establishments, and even businesses, to whom $10k is budgetary "rounding error"; no, it wouldn't be easy to locate the funds, but once tapped into it could exceed what you'd see from raw donations. There are also some businesses for whom Wikipedia would actually *enhance* their core product. Just to pull one example out of the hat: google.com. I notice that when I search on some of the terms I read about in Wikipedia (for more information), Wikipedia itself pops up as an important source. For example, 'bulldogging' or 'Gladstone Oregon'. Eventually (within a year or two), wikipedia will be filled with useful articles and will be a reliable source of information (compared with ad-laden commercial sites), and it would be in google's best interests to support and promote continued development (in much the same way that RedHat hires Linux Kernel programmers.)

You would, of course, need to be very clear in laying out the principles and processes that Wikipedia follows, and carefully elucidate the editorial control situation, and why it works in spite of (or even *because of*) the lack of central control. You will want to and need to stay true to the principles you've learned in operating the site, even if it means having to turn down some grants. Truly, the "proof is in the pudding" here.  :-)

-- BryceHarrington

---

I am a recent slashdot inductee, and I have to say first of all that this site seems to me to be a fulfillment of the real promise of the internet, which is not to make megabucks for corporations, but to facilitate communication and sharing of ideas, and to make the world a better place in the process. This Wikipedia somehow has hit the sweet spot, gathering relatively high (though uneven and sometimes not well organized) quality content from a large group of people and (we hope) ratcheting that quality ever higher. I think it is fine to make some money to pay a few salaries and keep the lights on, but I don't see how this can be made into a business first of all, and second if it were I think the contributions would stop or change in nature and the project would stall and die.

In real life I am a programmer coding on the free Linux OS, so I have seen how the internet can organically produce astonishingly high quality results (I hope this doesn't seem like unseemly advocacy, and I am aware of its flaws, I bring this up as an example of a successful internet project which it undeniably is) by harnessing the unpaid efforts of many contributors. One of the things that attracts these contributors is the guarantee of freedom to use, modify and redistribute the results of the shared labor.

I suppose I am preaching to the choir here, but I really would hate to see such a promising project fail due to a misguided effort to make money which ends up killing the goose. I second most of what BryceHarrington says above.

Jmccann

---

Knowing both myself and Jimbo, I can guarantee that we would never do anything that would "end up killing the goose." But I disagree with Bryce about making Wikipedia a non-profit. A non-profit couldn't support it, for one thing.

Pray tell, Larry, why?

Indeed, what is your basis for saying such a thing? Non-profit organizations support many very extensive projects, including ones that are vastly more complicated and expensive than Wikipedia. Just one example off the top of my head is the Canadian Arthritis Society, to which I was formerly employed. The number of interelated projects supported by this non-profit is simply staggering. I don't think that Bryce's suggestion should be written off by unsupported one-liners. -- STG

Besides, it's the old win-win situation of the public benefitting from a company's profit motive. Netscape did the same with ODP. I'm sorry if you don't like that, but it can't be helped. --LMS

---

And non-profits in the U.S. have rather stringent guidelines to go by in endorsing political candidates and views.

Really? What guidelines? That they must endorse, or must abstain from endorsing?

That is only true if it gained non-profit status from the IRS. They can still endorse views, but they can not candidates. So, a organization supporting a proposition on a direct-democracy ballot could (I think) be non-profit, but not one supporting Joe Blow for President. Or say a non-profit organization wants a presidential candidate to speak, they would have to bring in the opponent as well. I do not think that if Wikipedia tried to get non-profit status so people could make tax-deductable contributions the problem would be the IRS thinking we were political, but all the other stuff like having a Board of Directors and things. - Eean

I break into a cold sweat thinking how Larry would enforce the Neutral point of view then.  ;-) --KQ

---

I think banner ads will put off the exact type of people who are initially attracted to the project. All of a sudden they will think, rightly or wrongly, "Whoa, that dude is making money off of me!". And they're gone. If the authors are gone, readers alone do not yet generate enough page visits to sustain advertising income.

I also think that the Neutral Point of View does not allow advertising.

Why not apply for grants? There are tons of organizations out there which throw around money. A cool $150,000 could easily sustain the project for two years. And those are figures that are constantly given out for educational projects. You hack together a proposal over a weekend (containing the right keywords: access for the disadvantaged; empowering developing countries; collaborative; generating public property; bla bla), and then send it to fifty organizatons. One will bite. Many profs in academia hop from grant to grant on their little shitty educational software projects. --AxelBoldt