Project Wonderful Talk
The Unofficial Project Wonderful Blog and Forums
Contribute  :  Advanced Search  :  Site Statistics  :  Directory  :  Mass Bidding Software  :  PW Rapid Search  :  PW Selling Search  :  Links  :  Polls  :  Forum  :  DokuWiki  
   

Doing The Math

(contributed by Matthew Skala)

Project Wonderful's prices are low, but I don't think most users realize just how low they actually are. I did some comparisons between ads on PW sites and other similar ads sold through conventional channels, and the results are pretty clear: the same ad space sold through PW will command a much lower price. That's awesome for buyers, but it's not so awesome for sellers - which is why I'm listing this under the "Selling" topic. A site that could sell its ads some other way isn't going to want to switch to PW if it means accepting pennies on the dollar of the ad revenue it would otherwise get, and I think that's why PW hasn't become very popular with the big, high-traffic sites yet. Those sites often don't price things in terms of CPD like PW does, but for comparison I'm going to attempt to compute CPD for a range of sites. Let's see what we get.



First, let's look at one of my favourite Web comics, Megatokyo by Fred Gallagher. I'd classify it as a semi-pro site - originally an amateur project, and he still calls it a "doujinshi" (Japanese for amateur comic), but advertising and merchandise sales from the comic are the main income source not only for Fred, but also his wife and (if I remember correctly) at least one other full-time employee. I like Megatokyo as an example because it gives a fair bit of detail on its ad rates without demanding that you contact them for a "media kit".

Megatokyo has one 468x60 main banner and two 234x60 half-banners. They claim to get 350,000 page views on Mondays, Wednesdays, and Fridays (MWF), and 280,000 on the other days of the week. (They claim that MWF are "comic days," but that's optimistic - Fred is notorious for not meeting his own schedule.) Unique visitors are claimed to be 130,000 on MWF and 90,000 other days. The main banner costs $1.60 per thousand impressions. They say that will last a full day, but I think that means it'll stay in the rotation for a full day - you don't get exclusive use of the spot. Their claimed number of page views is 2.17 million per week, or 310,000 per day. Multiplied by $1.60 per thousand, that's a CPD of $496 for the main banner. For what it's worth, though, that banner space often seems to carry an ad for their own merchandise store, so they may be having trouble actually selling it at that price.

For comparison, the highest CPD (5-day average) quoted for a same-sized banner on Project Wonderful is currently $12.75 for White Ninja's main banner. White Ninja gets about 96,000 non-unique hits in a day, so it's just under one third the size of Megatokyo; but $12.75 is less than $496 by a factor of about 39. As for Megatokyo's half-banners, those go for $200 per week, and it's worth mentioning that Fred often has trouble selling them (one is empty right now). $200 per week is a CPD of $28.57. I can't find any half-banner ads on PW at the moment at all, so there's no direct comparison there. The final total for Megatokyo's non-PW ads is a CPD of $553.14.

Questionable Content by Jeph Jacques is another fine Web comic. They've got a 468x60 banner at the top of the page, managed by Indieclick. They also have four PW buttons at the bottom. I don't have current prices for Indieclick, but I have one of their "media kits" dated "Spring 2006", and at that time, they charged a CPM of $8 for Questionable Content and were claiming it got 4 million hits per month (if I'm reading the table correctly). PW says that Questionable Content gets just over 100,000 hits per day, which works out to 3 million per month.

Going with that figure (it's the more conservative choice, leading to a lower CPD), the banner ad at the top of Questionable Content has a CPD of $800. Wow! The four buttons at the bottom are (current PW bids) going for $4.10, $4.10, $4.30, and $4.10. Granted a button is not as valuable as a banner, and they're at the bottom of the page, but it's still a stark contrast. Jeph doesn't have a strong incentive to switch his banner to PW.

Diesel Sweeties by R. Stevens is part of the same collective with Questionable Content, Dinosaur Comics, White Ninja, and several other big names, and it has an Indieclick banner, too. Noting that Dinosaur Comics is drawn by Ryan North of Project Wonderful, I think we have him to thank for these comics being in PW at all - they're willing to participate in their buddy's project even though a purely economic analysis might argue against it. (I'm not including Dinosaur Comics in my comparison because I think it being Ryan's comic is likely to make it not a typical example.)

Again from the Spring 2006 Indieclick media kit, the CPM for Diesel Sweeties is $8. PW's traffic number for Diesel Sweeties is only 31,430 at the moment, which is interesting because the Indieclick media kit claims 4 million impressions per month, the same as Questionable Content. That throws the numbers into question - but if we use Indieclick's old price and PW's current traffic number, the same as we did with Questionable Content, the result is a CPD of $251.44 for the main banner. Diesel Sweeties also has six PW button ads, which are going for an average of $1.08 each at the moment.

I'm not going to trudge through the entire Indieclick/Dayfree Press lineup, but I will mention Kent Earle's White Ninja because it actually is running on PW advertising alone - NO Indieclick or similar - and it holds the top bid record. White Ninja has a 468x60 banner and two 117x30 buttons. Current top bids on those are $11.30 for the banner and $1.90 and $1.60 for the buttons. (The $12.75 I quoted earlier was the 5-day average, not the current bid as of Sunday night when I'm writing this.)

Let's look farther afield. Keenspot is home to a lot of the better amateur Web comics. You can buy Keenspot advertising at a CPM of $1.00. It seems to be one of these banner rotation networks, so CPD isn't so meaningful, and I'm not going to re-run all these calculations in terms of CPM instead of CPD - see my forum posting for some earlier thoughts on that - but you can compare the $1.00 for Keenspot to $1.60 for Megatokyo and $8 for Indieclick, as well as the fact that PW advertising is proportionally cheaper across the board. Indieclick also does advertising for a lot of other kinds of Web sites besides comics, such as the noted porn site SuicideGirls.com; the Web comic examples I've cited all had CPMs of $8, but the other sites mentioned in their media kit ranged from $5 to $9.

Since we're mostly Web comic people here, let's look at some other Web comic advertising venues. OnlineComics.Net is a big forum/listing site. One of their main advertising offerings is something called "sponsor listing", which mostly consists of preferred placement in their search engine; that's hard to compare, so I'll ignore it. The other thing they offer is currently six (I think they would add more if they had more buyers) 88x31 buttons for $19/week each, which is a CPD of $2.71. They claim 2.65 million impressions per week on that page, but I think that's a typo, because on another page they claim 2.65 million impressions per month, and that's more believable. Going with the per month figure, it's about 88,000 impressions per day. PW doesn't have 88x31 ad buttons, but for rough comparison purposes I searched for 117x30 buttons with between 75,000 and 100,000 impressions per day (5-day average), and the highest 5-day average bid value was for ComicSpace, happily enough a very similar kind of site, current 5-day average bid value listed as $3.30 and traffic as about 94,000. So that's an example where the highest PW CPD rate ($3.30) actually beats the non-PW rate ($2.71). The second-highest PW rate is White Ninja again with $1.70, so only the very highest-priced comparable PW site commands a higher price than what OnlineComics.net is asking without PW.

TheWebComicList.com is another popular listing site. They offer a skyscraper side-bar ad (I don't think it's exactly the same size as PW's skyscraper, but it seems to be a very similar thing) for a CPM of between $0.66 and $1.00 depending on how much you commit to buying. It's a rotation system instead of an exclusive ad, and they don't publish a hit count there, but on another page they claim to have "over 100,000 unique visitors" per month. The skyscraper rotation limits ad displays to three per unique visitor, so let's suppose they hit that limit, and have exactly 100,000 unique visitors (again, the conservative assumption to give as low a price as possible, as with the other sites), so they'll sell 300,000 impressions per month at between $0.66 and $1.00 CPM for a price per month of between $200 and $300. CPD is between $6.66 and $10.00.

OnlineComics.net also offers (they advertise as eight, but I count) ten 80x80 square buttons for "sponsor comics". Those cost $15 per month, so CPD is $0.50 if your button displays all the time, but it's another rotation system so you might not really get a full day's display for your $0.50.

I'd like to finish with a couple of amateur examples. I arranged a private placement with TA Vision (caution - nudity on the linked page), which is a small amateur comic and currently on semi-hiatus. To be honest, I viewed it partly as a donation to support the artist, so I wasn't too concerned about getting good exposure; but for about US$50 (I paid in Canadian dollars and don't remember the US exchange rate at the time), I got a year's placement of a 200x40 mini-banner. That's a CPD of about $0.14. I don't know the traffic numbers for TA Vision, but I get about as many clicks from that ad as from similar ads on PW sites with about 2000 hits per day. A CPD of $0.14 is a little on the high side, but within the range, of what similar-traffic sites get on PW.

Finally, there's my own Web site, which I think is typical of sites that might try to support themselves through PW. This one is interesting because I know how much it costs to run my site, and that's relevant to knowing whether sites can support themselves through PW. My main site is Ansuz and attracts about two thirds of the traffic - mostly search engine traffic because I provide a lot of information on legal and technical topics that people search for. The ads also run on my own Web comic, Bonobo Conspiracy, which accounts for about one third of the traffic. PW reports my current hits at about 2,500 per day, but I think that's higher than normal. There were a couple of spikes in the last few days because I ran ads with nudity in them and got much higher click-through than I expected; I doubt that will last once the novelty wears off and/or the sites I'm advertising on get upset and ban me, and the more usual traffic figure would be in the 1,600 to 1,800 range. I have a total of eight PW "button" slots; the current bids on them total to $0.13, an average of 1.6 cents per slot CPD.

What does it cost to run my site? What kind CPD would I need people to pay if I wanted to run a profit? I spend about Can$200 for my hosting account, and another Can$50 for registration of several domains. If we pretend those are the only costs (ignore the somewhat random advertising budget), that's about US$200 total per year (the exchange rate eats the $50). I need to cover US$0.55 of costs per day to break even. Since PW will take 25% of bids, my total bids have to be $0.73. If I stick with eight slots, the average CPD to break even is 9.1 cents; quite a lot more than the 1.6 cents I'm getting. And, I emphasize, that is just to break even.

So the bottom line is that I'm nowhere near being able to cover my hosting and domain costs with Project Wonderful advertising. I would need my bids to increase by a factor of almost six. Since I was content to run my site out-of-pocket as a hobby anyway before PW came along, whatever I make on PW is to some extent gravy - and I plow it back into buying ads on other sites, treating PW more as a link exchange with pennies used as game markers than as a profit-making venture. But that's a good thing, because if I were treating it as a profit-making venture, it would be somewhat less than awesome. You don't make a profit charging one sixth of your costs.

I don't know how much it costs the big commercial sites to operate, but I think the prices they quote (described above) give some idea of the scale. It's clear that PW doesn't pay anything like those prices. I think the big sites probably can't afford to sell PW advertising, certainly not on any basis other than "We like Ryan and will play his game, a little". I think that's why we aren't seeing the big-money sites participating in a serious way in PW - the prices are just too low.

Summary of my CPD estimates for quick comparison:
Megatokyo main banner: $496
Megatokyo half-banner: 2@ $28.57 each
White Ninja main banner (PW): $12.75 (5-day average) $11.30 (current)
White Ninja PW buttons: 2@ $1.75 each
Questionable Content main banner (Indieclick): $800
Questionable Content PW buttons: 4@ $4.15 each
Diesel Sweeties main banner (Indieclick): $251.44
Diesel Sweeties PW buttons: 4@ $1.08 each
Keenspot: CPM $1.00, rotation network, compare to Indieclick's $8.00 and Megatokyo's $1.60
OnlineComics.Net 88x31 partner buttons: 6@ $2.71 each
TheWebComicList.com sidebar: $6.66 to $10.00
TheWebComicList.com 80x80 square: 10@ $0.50 each (to be in the rotation; you don't get 100% display)
TA Vision odd-size mini-banners: 3@ $0.14, with space for at least one more
Ansuz/Bonobo Conspiracy PW buttons: 8@ 0.016
Ansuz/Bonobo Conspiracy break-even point: 8@ 0.091

Add to any service

Article Options

Trackback

Trackback URL for this entry: http://www.projectwonderfultalk.com/trackback.php/doing_the_math

No trackback comments for this entry.
Doing The Math | 6 comments | Create New Account
The following comments are owned by whomever posted them. This site is not responsible for what they say.
Doing The Math
Authored by: EddieC on Tuesday, January 16 2007 @ 03:00 AM GMT

Great analysis... thanks!

Here's another perspective from the other side, as an ad buyer:

I've been buying ads from WebbedComics.com for a long time because I wanted to support the site (and to promote my comic). When the site switched over to PW, I decided I wanted to bid high enough so I would end up paying about the same as I used to, so I could continue to support the site at the same level. I couldn't have done that without the other bidders playing along, helping to elevate the price.

In other words, I can't place a MINIMUM bid. (And the site owners can't require a MINIMUM bid, like on eBay where the seller sets the starting price)

If minimums were allowed, the PW prices would rise.

Eddie
Doing The Math
Authored by: mskala on Tuesday, January 16 2007 @ 08:26 PM GMT

Yes, the idea of letting sites set their own minimum bids is one I mentioned in my forum posting which inspired this article. I don't know how well it would work, because at least at first, sites with minimum bids just wouldn't get bids, period. Advertisers would rather go where they can get a bargain. But if minimum bids are what it takes to get the big-time sites participating, maybe after a while advertisers would start being willing to consider paying the minimum in order to get exposure on those big-time sites. A penny ad on "Freddy's Web Barn" costs less than $10 worth of advertising on Megatokyo, but it may not be a bargain when you factor in quality of the exposure.

Just today I had an experience with a site I won't name that is trying to enforce a minimum bid in the current system. I placed my free bid, had it rejected, and on digging around in trying to figure out why, I finally found a comment included in the advertiser's user profile (NOT in the description of the ad box) saying that they would reject zero bids. So I placed a penny bid, had it accepted, and now I'm still not paying anything because there are only just enough bidders to fill the box. If they get another bidder and my cost goes above zero, I plan to cancel the bid. My general feeling is to be offended. Who the ---- do they think they are to try to opt out of letting the market decide whether their ad box is worth paying for, when all the rest of us are taking our chances with the auctions? It's not even like their site is so desirable anyway, with only a few hundred hits per day. If they weren't so rude, I'd be more inclined to give them a chance with a real non-zero bid. But on the other hand, I know how hard it is to get a fair price for advertising, and I can understand the frustration of seeing only null bids for a box you know is worth more. And note that even with this site's "we reject zero bids" policy, they still can't really force people to pay, because they need to get enough non-zero bids to more than fill the box before those bids will start paying. I'm getting free advertising on there right now, policy notwithstanding. So far they don't seem to have ever made a profit, and as long as they stick to their current policy which alienates advertisers, I don't think they will.

One idea I've proposed in email with Ryan is that maybe the top bidder should pay their *own* bid instead of the current system of "top bidder pays just enough to beat second bidder". If we had top-bidder-pays-own-bid, then A. you would know exactly how much you need to bid in order to be assured of actually winning (beat the current bidders, instead of "guess how much they're bidding and then hope to bid more than that"); B. the cost of an ad would be more predictable, instead of the current situation where your bids can increase at any time until they hit their maximum; C. it would eliminate some predatory bidding practices, such as placing a bid you don't intend to pay out on just to force someone else's up; and D. (the point you raise) it would be easy to pay a site more than you're required to pay, if you want to express your support.

Letting ad box owners bid on their own boxes - currently forbidden by the rules - might be worth considering, too. It would be almost the same as letting them set a minimum bid requirement, and it would let them have even more control over what appears in those otherwise-unused slots. Other auctions, such as eBay, allow sellers to set a reserve price and/or bid in their own auctions.

Doing The Math
Authored by: Abalone on Wednesday, January 17 2007 @ 06:57 PM GMT
It might be good to have a "minimum bid" setting that replaces the ad content with other HTML content of the site owner's choice (i.e. an ad of their own, or some Google ads) while the bid hasn't been met, but still with some Project Wonderful branding underneath: "Replace this stuff with your ad: $x.xx", where that's the value of the minimum bid setting.

This would mean that people who were happy enough with their current advertising schemes would have a reason to get involved with Project Wonderful, which they don't at the moment - if you're getting a fairly steady $10 a day from AdSense, there'd be some benefit in bolting on a "Replace this stuff with your ad: $12.00" Project Wonderful box, so that if anyone really wanted to advertise there, they could, without any particular book-keeping effort on the part of the site hoster.
Doing The Math
Authored by: DiabloD3 on Monday, January 22 2007 @ 01:03 PM GMT
Actually, a "use this html when under $x bid" sort of feature would work really well. As it stands, I use iframes to rotate ads in, which is really screwing up PW's ability to collect "who referred you to this site" statistics, and being able to have PW to be featured directly on pages would help a lot in this area (as the tracking bug would be on the pages themselves, not in the frames).

However, a "use this html when under $x bid" by itself won't work, you need one additional feature: it needs to slowly phase in PW ads as it approaches $x, ie, if your minimum bid is $1.00, and it reaches $0.25 it does 25% exposure, if it reaches $0.50, it does 50%, if it reaches $1.00, 100% exposure. If you don't have this, then people won't place bids because low bids won't get displayed at all.
Doing The Math
Authored by: Kiba on Wednesday, January 17 2007 @ 12:45 AM GMT
Good article.

However, I could expect price to raise dramatically as more advertisers get on into the system.

As long the number of advertisers increase faster then the number of sellers, I expect to see higher price.
Doing The Math
Authored by: Caveman Joe on Wednesday, January 17 2007 @ 01:01 AM GMT
I agree, albeit with some reservations. You're right in that the price will increase if more advertisers join - however, the question of whether we'll see more advertisers than publishers is a difficult one to answer, and leads to a sort of chicken-and-egg debate. Also difficult is predicting just how high the prices will rise.
And I'd also add "as long as publishers don't flood their sites with lots and lots of seperate ad boxes" to that. ;)
The only thing I'm relatively sure of so far is that the economy of Project Wonderful will continue to rise and fall in waves (like we've seen recently) for a while before it even begins to stabilise.