A month ago I set up a head-to-head comparison on some ad banners for Leif & Thorn, and figured other queer-webcomic creators might be interested in the results.
The question is, does it actually entice more people to click on your m/m comic ad if it says “BL” in the corner? Or could that label be a turnoff, since some people have bad associations? What about a more neutral label like “LGBT+”, does that help or hurt?
(…in case anyone doesn’t know, Boys Love is the term used in Japan for m/m comics, so it’s geared to appeal to the manga crowd. You mostly see it on ads for webtoons.)
So I threw together a Project Wonderful campaign and let it run. I’ll explain all the terms for readers who aren’t in the ad game. Or you can skip straight to the end for the final data.
All The Terms, Explained
A campaign is when you pick some search criteria, and then set PW to automatically bid on sites matching those criteria.
In this case, I focused on sites in the “Comics” and “Webcomics” categories. (People reading the sites in “Home and Garden” or “Business and Finance” are…probably not going to click a lot of webcomic ads.)
You can also narrow by the specific keywords people have put on their sites — which gets great results if, for example, you make an ad featuring your vampire characters and splash it over sites tagged “vampires” — but for this I kept it simple.
Views and clicks are exactly what they sound like. PW also tracks uniques — that is, if one reader binges the entire Leif & Thorn archive, then whoever has the winning bid on the skyscraper ad will see their stats increase by 926 views, but only 1 unique.
CPM is cost per million views. The lower it is, the better you’re doing at making cost-effective bids. (It goes by the ratio, whether or not you’ve actually gotten that million. If you spend $1 running an ad that gets only 1 view, then the CPM is $1,000,000.)
CPC is cost per click. The lower it is, the better you did, because it’s enticing more people to click on it for the same price. It’s a factor of both “did you make a good ad?” and “did you do a good job picking sites with audiences the ad will appeal to?”
CTR is click-through rate. The higher it is, the better, again on both “the ad is good” and “the sites are your target audience.” But don’t get too excited — it’s going to be less than 1%, probably less than 0.1%.
So I did this with two different ad images. Both come off as pretty romantic — one fluffier, the other more dramatic. Same caption for everything — “Love in a second language.” (In an earlier caption-based faceoff, that line did well.)
There are 3 versions of each ad. One blank, one with “BL” in the corner, and one with “LGBT+” in the corner.
And here, after running all of them in equal measure for a month, are screenshots of their stats. Click for full-size:
I checked in several times over the course of the month, and sometimes one label or the other appeared to be pulling ahead, but over the long run? Absolutely no difference.
A logical follow-up would be to see whether the labels help (or hurt) on banners where the image isn’t already telegraphing “hey, this is pretty gay” on its own. And/or to see what effect they have in a campaign that targets sites with LGBT-related tags, where you know it’s a specific point of interest with the audience.
In the meantime…at least I learned that “comforting backlit kissing” performs consistently better than “holding hands in snow.”