Image Ad Blending Works Really, Really Well

So I’ve hung around with SEOs for the last couple of years, including ones who pay the rent based on their ability to convince people to click on AdSense ads, and I’ve learned a trick or three.  One that Google will actually tell you in as many words is to make the ads seem less garishly out of place on your site.  In the words of an Internet buddy, “You want to not look like an ad at the first glance, but to look like an ad on the second glance.”  This way you avoid banner blindness, the phenomenon in which Internet users mentally tune out portions of a website which look like advertisements.

Google says:

We’ve outlined a few strategies below that are designed to decrease ad blindness, the tendency for users to ignore anything that’s separate from the main content of your site. By making these changes, you’ll be making your ads more visible to users. The goal isn’t to confuse users into thinking ads are content, but to get users to see and read the ads so they can click on those that interest them.

That’s why if you buy ads on Google itself, they’ll look something like this:

As you can see, the top ad is styled to resemble the main content on the page, with a bit of a subtle yellow background and the notation that it is an ad there, if you take a gander for it.  (Note that a lot of users think that yellow is used for marking the best results.  Non-technical customers, yay.  If you don’t believe me, watch someone who can’t program Google sometime.)

I buy a lot of ads on the Google Content Network, but I don’t get a say in how they are presented usually.  Since I pay per click, it is in the interests of webmasters to make the ads look content-esque, so that they catch lots of clicks.  They occasionally get clicks with less-than-true-user-intent volition behind them, but that is a cost of doing business for me.

Anyhow, I’m always experimenting with different ways to advertise profitably for my businesses other than Google AdWords, which a) are expensive (I spend something like 50~60% of my gross on a sale at the margin) and b) have limited volume.  I recently was inspired to try something new when listening to a (paid) Mixergy video with Ilya of Mixrank, whose blog you should really be reading if you are interested in online advertising.  The gist of the video was to try negotiating direct deals with advertisers with access to the right targeted demographics, and I’m going down that route as well, but for the moment I wanted a get-my-feet-wet option that was more self-service, so I went with BuySellAds.

BuySellAds basically lets you pick a website (in their network), pick a particular type of image ad inventory on it, and pay the displayed rate for advertising there for a month.  Sadly, their options for inventory appropriate to teachers are rather lacking, but I found one wonderful website accepting ad placements through them: BusyTeacher.  BusyTeacher takes 728×90 image ads on their category pages, like this one.  So I went into Paint.NET to exercise my meager pixel-pushing skills, slapped something together, and submitted it for their approval.  After they approved the ad, my credit card got charged for a month of placement ($135 for an estimated 500k impressions or thereabouts), and it went live.

Let’s Play “Spot That Ad”

You can click on the photo to see the full size version.


You probably saw-but-didn’t-see the “Get A College Degree For Easy Loan Payments of Only…” spammy ad in the middle.  If you routinely surf sites visited by middle-aged women, you’ve seen-but-not-seen so many thousands of them that you tune them out.  But you probably didn’t automatically tune out the Bingo Card Creator ad.

Spot it yet?  Hint: it’s the row without the Facebook button.

Is This Evil Or Just Evil Genius?

Once upon a time I was an engineer totally scornful of effective marketing, but I have gradually gotten over it.  After thinking it over, this is aggressive but within my comfort envelope.  The ad is honest about being an ad, makes a straightforward commercial proposition (“Sign up for a free trial”) to an audience that I think will respond well to that, and is pretty true by the standards of marketing copy.  It is designed to catch clicks only from people interested in signing up for a free trial of Bingo Card Creator, and sends them straight to a landing page where they can do just that.

I wish there was a way to dynamically generate the image such that I could provide a more exact star valuation, but in the context of a sponsored placement, “Rated 5 stars by lots” is both non-specific and true.  Lots of people have used BCC, and when I ask for star ratings in internal surveys I get something like 4.8 on a volume of hundreds or thousands.  I think this compares favorably with “9 out of 10 dentists agree” and other pretty banal marketing copy.

So Does It Work?

Oh heeeeeeeck yes.  That ad has a 2.2% click through rate (astronomical by the standards of banner advertising), generates about 500 views of my landing page a day, and of that about 11% or so convert to the free trial.  (This is lower than my landing page average by quite a bit, but that makes sense.  Most people who come to one of my landing pages just got done searching for e.g. “make bingo cards”, so they’re clearly in a bingo mood.  The user here was just looking at a page about generic teaching activities then shown the bingo option, and might not be totally sold on bingo for her classroom yet.)

Anyhow, at about $5 to run on this site per day, the CPA (cost per action = how much money do I spend per free trial signup driven) has been running somewhere in the 8 cent region.  Yowza.  I pay Google closer to 30 ~ 40 cents.  At typical conversion rates to a trial of about 1.8~2%, that means that this costs me $4 or so to generate a $30 sale rather than $20.  Sold!

I just wish there was more inventory available.  Many of the sites in the teaching niche either only do Google AdWords, so I’m already saturating them (and paying a cut to Google), or they only accept advertising through large networks, which tend to favor e.g. brands with $X00k marketing spends and not guys who want to experiment with a few hundred bucks at a time.  I’m following up on the advice to get in touch with smaller sites directly, but I need to hit the sweet spot of “Small enough for my $X00 to matter, large enough that they send me enough traffic such that my time in negotiating an ad buy and preparing a creative is worthwhile.”  For an experiment that looks like it will net in the neighborhood of $1,000 a month in sales for $135 and ~10 minutes of pixel pushing, this one is going in the win column.  I hope to get bigger and better results later.

Anyhow, check out BuySellAds, direct ad buys, and ad blending.  They can be made to work.

About Patrick

Patrick is co-founded Starfighter, founded Appointment Reminder and Bingo Card Creator, and presently works at Stripe on Atlas. (Opinions on this blog are his own.) Want to read more stuff by him? You should probably try this blog's Greatest Hits, which has a few dozen of his best articles categorized and ready to read. Or you could mosey on over to Hacker News and look for patio11 -- he spends an unhealthy amount of time there.

34 Responses to “Image Ad Blending Works Really, Really Well”

  1. luke November 23, 2011 at 10:43 am #

    Haha, I didn’t know what you were talking about because I use adblock plus so I never see ads.

  2. Anonymous November 23, 2011 at 10:54 am #

    That’s not yellow.

  3. Joe's Pferde Bilder November 23, 2011 at 11:07 am #

    I had also to use an extra browser (safari) where adblock is not installed to see what you were talking about. But ad blending seems to be a good way to convert clicks. (:

  4. john smith November 23, 2011 at 11:08 am #

    SEO scum

  5. Daniel November 23, 2011 at 11:09 am #

    I’ve used this as an effective ad strategy for a while:

    1) Run Google Content ads. Use generic keywords to get a lot of impressions.
    2) Use Google’s Placement Report to find out what specific sites are converting customers.
    3) Contact those sites directly and offer to buy out their entire ad inventory. Pitch something like “I’ll pay 115% of whatever you make off AdSense if you remove all those ads and replace them with mine. We’ll run a one month trial and see what you think…”
    4) Design ads to fit in with the content of the site.

    Webmasters love this because AdSense doesn’t pay much. And I get a lock on a great marketing channel.

    The hardest part is getting a hold of the webmaster. And many are reluctant to believe you are not a scam. Some of the sites are just some guy’s blog that happen to rank well for a specific phrase. Guaranteed cash usually helps (and is worth the risk on our end).

  6. jes5199 November 23, 2011 at 11:20 am #

    What’s really amazing is that adblock blocked your screenshots on this page

  7. seanp2k November 23, 2011 at 11:24 am #

    You know what happens when I click an ad like this and expect good content on the other side (which is very very very rare — otherwise, why would you /pay/ for me to click on it?)? I never visit that site again.

    Good job polluting the internet with confusing crap that people don’t want.

    Why not actually spend time making a real store selling things that people actually want to buy instead of playing these games trying to trick people into clicking your crap? Yeah, you can make a career out of being an asshat like that, but is that something you can be truly proud of?

    You are the modern-day equivalent of the shell game.

  8. twmb November 23, 2011 at 12:10 pm #

    I recommend taking the “ad” out of the image names so that people don’t have to turn off ad-block pro or plus or anything to view the images.


  9. khromov November 23, 2011 at 12:23 pm #

    This is very interesting. Thanks for this post.

    I think it’s easy to feel cheated, but only if the content you click doesn’t match your expectations. In essence, the AdWords banners (which most people tune out by default) are generally of low quality, but the kind of ads Daniel talks about where people cooperate with the site owners to bring relevant content is refreshing as opposed to AdWords farming.

  10. Jay Caines November 23, 2011 at 12:49 pm #

    I’d just like to say that I never had a lot of respect for you, and you’ve convinced me that my read was right. You make the world a shittier place.

    I wish the FTC was better about enforcement actions against cretins like you, because honestly, you should be paying a huge fine for that bullshit.

    Its disgusting that you’re proud of what a shitty, awful person you’ve become. Your mother should be ashamed.

  11. Woolwit November 23, 2011 at 1:26 pm #

    I’ve been hearing about these new img ads and concerned that they’d outsmart my ad blocker, AdBlock Plus (totally free browser plug-in). But I don’t see any of the images or ads you’re talking about in this post. It’s like you’re talking about nothing.
    The reader-as-ad-target internet paradigm is demeaning and humiliating. If your life is about manipulating people to click on ads so you make money, well, so be it. Have fun in Vegas.
    Get a button.

  12. kikito November 23, 2011 at 2:13 pm #

    Adblock-plus user here. I would recommend switching to plain images instead of hardlinking the ads.

  13. Engineer November 23, 2011 at 2:20 pm #

    Thanks very much for this article! It really helps to see an example of a working strategy for a business like yours.

    I was dismayed, but not at all surprised, to see the universally hostile reaction this article got on Hacker News.

    This confirms for me that Hacker News is inhabited by a bunch of kids who like the idea of being rich, but aren’t really interested in doing a startup, and certainly at a core level think that making money is evil and capitalism is bad.

    I think it would be really useful if there were a forum (rather than a reddit clone) where smaller startups and small team software and internet businesses could meet, network and share techniques.

    I’d love to share the stats on what we’re working on, and my experience. But HN is not the place, because there are too many stories, the stories are voted on based on the idle interest of the readership, and they aren’t’ really targeting helping people get stuff done.

    I’ve started a dozen businesses, but there’s always more things to learn. I’d love to share this knowledge somewhere more intimate than a blog (some stats you just don’t want to talk about) and where you can do so without being attacked by the kind of ideologues who inhibit HN.

    If you know of such a place, please send me an invite and tell me the secret handshake!

    I’d hate to do “Yet another” type thing like this, but I’ll build my own if I have to.

  14. Bjoern November 23, 2011 at 3:33 pm #

    This is great stuff. I don’t see any moral issues with this like some of the ad-hominem insult slinging commenters before me. As long as someone puts in the work to find relevant audiences I don’t see anything wrong with it. If the content owner deems the ad relevant and is willing to put his/her reputation on the line what’s the problem?

  15. Elissa P. November 23, 2011 at 3:45 pm #

    I think that the key is to make the ad look just like the content. I’m also going to try your ad buying plan in order to create sales at my website. I think that I can get a better conversion than I’m getting now.

  16. anonymouse November 23, 2011 at 4:56 pm #

    It is yellow. Just the wrong shade.

  17. Robert Seaton November 23, 2011 at 5:00 pm #

    For what it’s worth, Patrick, I think that most of the reaction to this has been totally overblow. This is evil? Really?

    I’m pretty okay with this. It’s probably a little grey hat, but as long as the webmaster thinks it’s okay, whatever works, right?

    If you haven’t done this already, you might want to try to get some seasoned affiliates selling your product by joining an affiliate network. If you offer a decent commission, I’m sure you’ll see sales.

  18. Adam November 23, 2011 at 5:06 pm #

    Way back in the day, I did some banner advertising on Altavista. I followed the same approach where my ad looked almost exactly like the native search results; like the number one result. Back then, they had this crazy notion that an ad had to stand out and blind you, with crazy flashing colors and other usability non-sense. My ad was simple, clean, and like you, extremely effective. Perhaps it was a little too deceptive, but even back then (before adblock), people very quickly learned to ignore ads.

  19. Jason Fuerstenberg November 23, 2011 at 8:50 pm #

    A lot of people hate ads but advertising in and of itself is not evil.
    If you cured cancer and never advertised the fact nobody would stand to benefit.

    But I agree that some ads are as you called them, “spammy”, and that is a fault of the marketing team that created it.

    Having said all this I don’t advertise my own product except via PRs through companies like prMac which tend to be really affordable and get wide distribution for a few weeks.

  20. Lindsay November 23, 2011 at 8:59 pm #

    “I’m pretty okay with this. It’s probably a little grey hat, but as long as the webmaster thinks it’s okay, whatever works, right?”

    Riiiight. Because if *two* people are complicit in a greedy and dishonest tactic, and it works, that makes it less evil.

    What the hell, people.

  21. Cameron Elliott November 23, 2011 at 9:29 pm #


    Thanks for this posting, I hope the ninnys go away and don’t discourage you from sharing such insightful, and quantitative posts about real online business.
    Your blog is golden, please stay true and forthright.


  22. Lindsay November 23, 2011 at 9:30 pm #

    This is an issue of advertising ethics which is decades, if not centuries, old. Anyone doing even a rudimentary marketing or business course would have learned why this is unethical, and a bad business decision *for the publisher*.

    Newspapers and magazines, another medium almost entirely dependent on advertising revenue, have internal policies in place to prevent this kind of scam marketing. This is primarily to protect the publication itself from loss of the trust of their readership, based on the fact that this is clearly a dishonest tactic which would rightful earn them the scorn of the people who they wish to retain as “users”.

    See the American Society of Magazine Editors guidelines, section 3:

    Yes, some editors are more unscrupulous (read: greedy) than others, but as above, having two parties complicit to an unethical decision does not make it ethical. It’s called “collusion”.

  23. raul November 23, 2011 at 9:52 pm #

    Patrick, thanks for the share. It’s really an interesting view on adds/advertising.

  24. November 24, 2011 at 1:41 am #

    So what you’re saying is… making your ads blend in with and actually relate to the content, rather than being annoying flashing contrasting graphics distracting you from it, is a good idea? What a revolutionary concept.

    Seriously, I can’t believe this isn’t too obvious to write about.

  25. Peteris November 24, 2011 at 3:00 am #

    I work for a company that runs a lot of websites, and the one major rule for any ads is that ads cannot reuse any UI / style elements of the site or in any other way be confused with the site content (such as your example with the ratings stars, fonts, colors).
    So any such ads would be immediately rejected w/o any discussions.

  26. kosh November 24, 2011 at 3:11 am #

    This activity is illegal in many countries, by crossing the line into deception. Your ad should never have been accepted.

  27. Melbear November 24, 2011 at 3:58 am #

    “This activity is illegal in many countries, by crossing the line into deception. Your ad should never have been accepted.”

    In what way is this illegal? I think it’s great marketing.

  28. informavorette November 24, 2011 at 5:36 am #

    You say that you think this is ethical, because1) the add doesn’t hide its being an add, and 2) it leads people interested in a teaching tool to a teaching tool. I disagree with you.

    The first point: it does have minor differences from non-sponsored content, but the customer has to pay close attention to details to notice it. The ad mimics content closely, and I can imagine that many people clicking it didn’t notice that it is an ad (the “astronomic” click-through rate supports that – if a non-mimicking banner also leads customers to relevant content, why does it have such a lower click-through rate)? So, from the POV of a customer, your ad is a trap for the unwary. When a customer notices that, he is distrustful of both you and the publishing site, as Lindsay pointed out. So you mislead customers who don’t notice they are clicking on an ad and alienate the rest.

    Your second point is that, whether a person is attentive enought to notice that this is an ad or not, they were searching for a teaching tool, and your ad took them to a site for a teaching tool, so they should be happy. But these customers didn’t visit the aggregator site to get anything parading under the name “teaching tool”. The reason customers visit aggregator sites is to get an overview of what is available on a market they don’t know well, and to see how people more knowledgeable about the market rate the different offers. The aggregator site is an intermediary who provides information about the market to the customer.

    Placing an ad like yours seriously reduces the value of the aggregator’s service for the customer. The customer trusts that a fair aggregator will offer an unbiased view of the market. He would offer some kind of filter function – only placing tools worth buying on the list – and sorting function, by using a fair algorithm to place better tools in prominent positions. If some of the positions on the list are sponsored, the list is first not filtered – because the aggregator did not ensure that the tool has enough quality to be listed, or, if he did, there is no way to make that credible because he has a monetary incentive not to. Also, the list is no longer ordered by a fair, transparent algorithm, but some prominent slots are reserved for tools which did not compete with the others for a prominent placement but paid for the slot. They can no longer get the information needed to form an unbiased opinion on the tools offered on the market, because they are being fed biased information.

    This is why customers dislike such ads a lot – probably even more than flashing, rotating banners. They are informationally overloaded, and visit an aggregator site to reduce the amount of cognitive work they need in order to perform a task. Stealthy ads increase the amount of cognitive work needed and render the aggregator next to useless. The customers feel cheated and unfairly treated – even though your ad would indeed lead them to a tool which might be exactly what they need. And they feel very strong about this – you can see this plainly from the tone of the commenters who called you evil scum, etc.

    You pointed out that Google is doing something similar. But Google’s ads are not as deceptive as yours. First, the yellow background makes them easy to distinguish with a quick glance. Second, they all appear in their own area, above the other search results and not intermingled with them. So anybody who wants to ignore them can do so with minimal cognitive load. Yes, the first time somebody sees a Google SERP, they have to orient themselves and notice the small hint that the yellow results are ads, just as with your ad. But the big difference is that everybody on the Internet is familiar with the Google interface, while specialized aggregator sites like the ones in your example are a place the customer is likely to visit when doing his first research on a market new to him, so he doesn’t know their interface. And of course, it is much more difficult to spot your ad even after he is familiar with the interface.

    Bottomline: Advertising is not evil in itself, but deceptive advertising is unethical. Your argument “it takes people who want a teaching tool to a site selling a teaching tool” is wrong, because the customers aren’t looking for a teaching tool at that stage, they are looking for an unbiased list of available teaching tools, and you, together with the aggregator, are offering them a biased list of available teaching tools. This makes your strategy deceptive, thus unethical.

  29. Dennis Reinhardt November 24, 2011 at 8:59 am #


    There is a forum where startups and small ISVs can discuss the nuts and bolts of building a business: This is the ASP (Association of Software Professionals). ASP costs $100/year and that does a lot to confine the some 800 members to those serious about building a business. Supporting members pay $250/year (I am a supporting member).

    You should also know that the ASP sponsors a yearly conference: and that Patrick is one of our featured speakers.

  30. Alan November 24, 2011 at 6:09 pm #

    Cheating is for Google only!!!!

    Only Google can run ads similar to text and fill the page with ads.

  31. samuel November 25, 2011 at 5:24 am #

    Clever. I still have adblock. And if I ever land on a page with stealthy ads (that circumvent adblock), I’ll make a mental note of never going there again, because it’s just too tedious to use.

  32. Noah November 25, 2011 at 8:09 am #

    This problem has been around since the dawn of publishing. In the end, it is about quality, not ethics.

    How much do you, a publisher, value quality? How much are you willing to compromise quality to increase your ad revenue?

    In places of learning–libraries, Wikipedia, etc.–we do not want ads at all, so we can spare our users from distraction. In a commercial publication, we cannot afford that level of purity–but we can still insist on a standard of quality.

    Ads that are hard to distinguish from content are a sign of poor quality in a publication. Just like ads that pop up, move around, and flash.

    Let’s not get all worked up, people. Poor-quality goods will always have a market. Don’t be on a high horse.

    Try not to condemn people for choosing to compromise on quality. If you feel strongly about quality, then strive to put quality into everything you do. Then you lead by example.


  1. Ad blending is tricking us all | Josh Kerr - November 23, 2011

    […] you’ll get a significantly higher click through rate.  Patrick from Kalzumeus wrote a great article discussing the SEO behind the Adsense ads.  Here is an example from Google […]

  2. Image Ad Blending | studyo - November 23, 2011

    […] lita inte på mig helt och hållet, ta gärna del av den här trevliga berättelsen av Patrick McKenzie om hur bra image ad blending har fungerat för honom. Sök […]