ETA 2018: Project Wonderful canceled my ad box a while back because they thought it didn’t make me enough money. I always meant to get back into the game but as it turns out, they’re canceling all other ad boxes in August anyway. So consider this post a general write-up for potentially long-tail firendly ad services. If there even is such a thing anymore, after PW.
For the longest time, I’ve been running ads on this site. They were from Project Wonderful (PW). In case you don’t know it, that’s a comic-centric web ad system, the only one I’m aware of. Other than most ad systems I know, it doesn’t pay by page views or clicks, but for the time the ad is placed on your site, based on a system of bids so the pay depends on how popular your site is with advertisers. For small sites, revenue isn’t high, if any, but it’s good for a cent or two where other services won’t even bother to look at you.
About two weeks ago, PW introduced “geotargeting”. European users will now only see the ads targeted at Europe, US Americans will see US ads, Canadians get their own as well, while the rest of the world gets served bulk ware. Advertisers can direct ads to each of the four regions independently. According to PW’s blog Blogject Wonderful (BW), publishers (=site owners) can earn more by selling the same ad space four times at once, while advertisers can save money by having less competition to outbid in each region. Sounds good, doesn’t it?
Well, that depends.
In BW’s example, a site formerly getting $100 per day can make the same amount from $25 bids because it has four times the ad space to sell. That’s good. But if you’re at the bottom of the revenue pile, struggling to rise above the $0.01 mark every day, things just got harder.
You see, it goes like this: The first bid on any ad space will be counted as $0.0 because there’s nobody to outbid. The second bid will register at $0.01, and so on. Now if you have four ad spaces instead of one, you get four first bidders. If these four are all you get, your revenue will be zero as opposed to the $0.03 you’d get from four competing bidders.
Of course, it doesn’t work exactly like that because the four bidders may bid for the ad space in the same region while the other three got unbid. At my site, I didn’t suffer much revenue loss so far because the American bids are behaving pretty much the way they used to. Whereas bids from other regions have yet to pick up, mostly. (Looks pretty embarrassing from here, btw, with that substitute ad up all the time.) But it’s only been two weeks – a lot of advertisers haven’t tuned in to the new system yet. Once they do, I expect lower bids altogether (= little to no revenue).
Does it work for advertisers, then? Any system that suddenly drops prices up to 75% must be a good one, right? Again, that depends. With PW being as US-heavy as it is, that’s where most of the money is exchanged – I don’t see much of a change there. I expect the others to pick up as well, but so far they haven’t.
But who wants to restrict their ads to special regions anyway? Unless you’re selling physical goods and don’t care much about International shipping costs? What’s the point in targeting regions?
I’ve made a serious effort of using the new feature to my advantage. I’m still experimenting here, so I can’t give you a lot of numbers, but right now I’m using geotargeting as a very blunt timezone targeting system. USA and Canada are between GMT-5 and GMT-9 (or something like that), Europe is between GMT and GMT+4. For now, I’ve split all my campaigns into 2, one targeting USA & Canada during American afternoons, one targeting Europe during ours.
Since the other region includes Africa (European time zone, roughly), South America (AmeriCanada) and Asia (none of the above), I don’t really know yet how to make that one work. (Seriously, PW, “the rest of the world” is not a region!)
Additionally, I’ve placed some extra bids on whatever the cheapest region on popular sites was. (Believe me, reading Girl Genius is much more fun when every new page identifies you as a sponsor.)
These steps led to a geographic shift, but not necessarily to more traffic. European views picked up. A lot. Also, I’m suddenly getting a lot of traffic from Canada. (Good day and welcome, Canadians! Glad you could make it!) That’s probably because Canada is cheaper to advertise to than the US, so my US/Canada campaign wins more bids there. Fittingly, American traffic almost disappeared. Since my previous campaigns were targeted at all the world, but during American afternoon hours, that’s where most of my traffic used to come from.
For my purposes, actual timezone targeting would have been a much better new feature than country/another country/trade zone/rest targeting. (Another thing I’d like is language targeting so I could advertise my German site on other German sites. We’ve discussed that a lot within the German scene, hint hint…) But you use the instruments you have. Even if they’re blunt ones.
Doubling my bidding time means doubling my bidding expenses, of course. This is another feature that may work better on high-paying sites, if the winning bids really spread over the regions. But cheap campaigns can’t really get any cheaper, so more hours of cheap campaigns just mean more expenses. If I end up with more traffic now (I should be able to tell after a month or so), it’ll probably be because I’ve advertised (and paid) more, not because I’ve targeted better or spread the news worldwide.
Does that mean that PW’s geotargeting really just helps PW, as some critics have hinted? Not necessarily. My impression so far is: the more you’re making (or paying) already, the more likely you’re going to profit from the new system. Those of us at the bottom end, though, may well end up rolling even further down the Long Tail.