I was talking to one of the guys the other day about how the CPM on display ads for porn sites was far lower than you can get on Google We don’t run media on porn sites for any of our clients, but one day we may have a client that would want to get in front of the 624,000,000 daily viewers of the top 14 adult sites… that’s 20% of all Google daily searches. It’s a lot. Anyway, this post isn’t about advertising on adult sites, it’s about how some advertisers will intentionally create ads that are built to filter out people just by the ad quality alone.
Why you won’t trust a Nigerian Prince email
Most people who are reading this post are probably smart enough to spot a scam email; the standard Nigerian prince or World Bank deal. It smells fishy and you can tell it’s all bullshit. However, there is a method to this sort of ad, and yes, a spam email is an advertisement. The method is to use the content of the ad to filter out the possible enquiries. They are filtering out all the smart people, the people who are less likely to buy, or in this cane, less likely to get scammed, but the idea is solid.
Using the ad text is more than just coming up with something that will make people purchase. You can refine the quality of respondents just by making the text so bad, that only low educated people will respond. Spelling mistakes, grammar and an offer so outlandish that no reasonable person would contact you. Then you’re only going to get responses from dumb people. This means that the prince can send out billions of emails, completely un-targeted, and still only get inquiries from people that are more likely to convert.
That girl really does want to sleep with you, and she is in your area too!
This type of banner ad has always astounded me; the low quality graphic, the flashing pre-2000 gif style and the obvious (well maybe not that obvious), too good to be true, pitch. It just goes against everything that I believe is good quality advertising, BUT it works, it works very well. Again they are using these tactics very purposely, to filter out people that will not convert and that is just wasting money on clicks.
Now it starts to make sense, using that beyond the ad text or call to action, but it allows you to spread a message much further on a platform like display, using adult sites, without having to worry about targeting. There is no way to target a user on an adult site, beyond their kink, and I don’t think that you can really target a person’s purchasing intent on a computer, by their interest in Bored Housewives.
I do wonder if there is, or could be, a correlation between someone’s sexual interests and their purchasing demographics. It could be assumed that people that watch longer videos are more likely to be able to manage delayed gratification and that could possibly mean that they, through that ability to say “no”, are more likely to have a higher level of purchasing power because they can save money. It’s a big stretch and I may look further into the research on this and write a post If I find something of merit.
Making your ad text your filter so you can go broad
How do you take this understanding and apply it to marketing that isn’t in an adult network or a Spam email? It’s simple… Let’s say that you are trying to target a large potential audience in Facebook, in this case let’s say you’re looking to advertise to Bangkok users, 18-45 males only. It’s a very wide target so will likely blast your budget before you can make some real conversions. We could in most cases filter with job titles or education level or even interests, but let’s say that you either don’t know that graphic of your market or your customer base is varied and you are potentially going to lose custom with that targeting.
The simple way will be to use an image and that can be effective, especially if you have your own photo team. However, if you start using images of people in fancy cars to try to filter people with high purchasing power, you risk being just an aspirational ad and will still get clicks and traffic from users that won’t convert.
One way to improve ad copy is to add a price in. By stating a price in your ad it will filter out people that are less likely to purchase. Because you have a low price it is instantly compared to products that are similar and you may get that sale. Similar with high pricing, the customer may not click because it’s well above their ability to purchase.
Another way, and this has to be done with care, is to use slang in the ad copy. By using “street” language (and I am shopping my age here by calling it street) you can segment your market on that ethnographic metric. It’s taking the copy to another level of pitch; you are removing the “highly educated” from your traffic because they will “not talk like that” and it’s not going to connect. Beyond making a user feel like you are talking to them, we are using language to talk like them.
What’s the pay off?
At the end of the day your ad copy has many users. It’s to pitch your product to the best possible client, however, sometimes you don’t get to target your ads as well as you would like, so using the copy in the ads as your filter, you can spread your message much wider to a larger segment and still, with luck, only generate the most converting leads.
The down side is that most of your metrics will be low. Low relevancy scores and low CTR will have an adverse effect on your ads quality and running, but if sales are the only metric you want, then you can sacrifice these quality metrics. Also, you may run the risk of segmenting your brand too far and giving your brand a perceived persona based on your ad copy.
I think that in the end it’s not a bad strategy if you have no other options, but placement targeting, getting your ads to show only to the most relevant audience is always going to win.