Live data from Hacker News

Image Ad Blending Works Really, Really Well

kalzumeus.com

191–200 of 215 posts

Re: Image Ad Blending Works Really, Really Well

#191
post #59
post #52

Earlier quoted context omitted.

No. Only if Google included barely distinguishable paid placements in their organic results would it be a fair comparison. I'll concede that some minority of users might not be able to distinguish paid results from organic on Google, but no one could do it with these "blended" ads -- that's why he's getting such a high CTR. Users are being tricked.

The Google ad that I note-for-note copied has a CTR in excess of 20%. It should - it is exactly what the user was looking for, presented in a way that makes that obvious. [Edit to add: I thought I might have been exaggerating, so I looked it up. Nope, 21.3% this year for [bingo card creator]. For [bingo creator], which is not a brand term, it is 18.5%. [printable valentine's bingo card]? 8.7%. Do people get the pictu…

No, it's not "presented in a way that makes it obvious [that this is what the user is looking for]". It is pretending to be curated and endorsed by the site where the ad is shown, and it is displaying a fake rating pretending to be from the users of this site.

For comparison, Google ads are displayed in a right column by default, and you can't simply buy your way to the search results column. Google displays the ad there only when it considers the ad as relevant as a search result, and relevance judgement is something you already trusted Google for.

You are supposed to know this, aren't you a SEO / AdWords expert?

In Google's case it's also much more obvious that they are displaying an ad and not a regular search result. If you claim otherwise, why don't you copy the yellow background and "Ad - Why this ad?" text in your advertisement and tell us where the CTR goes?

I've seen you say that Google is evil and claim a moral high ground a number of times, and now you are justifying something with "but look, Google does it too"? Even if you were correct, which I've shown you aren't and you know it, this would means you are just as evil as Google.

Re: Image Ad Blending Works Really, Really Well

#192
post #91
post #56

Earlier quoted context omitted.

Let me clarify - I would much rather a teaching activities site be filled with BCC ads over weight loss ads. That doesn't justify the use of deceptive advertising beyond exploration. This is the -exact same- line of thinking that goes on at social games companies with viral game features. Good people say: "20% of users respond to this request, it must be interesting to people. Send more!" This is completely oblivious…

So, just so we can be as specific and absolute as possible: you also think that every Google SERP is dishonest. Because the whole premise of Patrick's post is, "here's how Google's ads work; I wonder what would happen if I crafted my ads to fit into pages the same way Google's do?"

The comparison to Google is incorrect, as I said elsewhere.

Google ads are displayed in the right column by default, and you can't simply buy your way to the search results column. Google displays the ad there only when it considers the ad as relevant as a search result, and relevance judgement is something you already trusted Google for.

Patio's ad is pretending to be curated and endorsed by the site where the ad is shown, and it is displaying a fake rating pretending to be from the users of this site. It's also way more similar to the non-ad text on that page.

Re: Image Ad Blending Works Really, Really Well

#193
No it doesn't. It's an awful, deceptive move.

You know why? It's because Google makes it easy to spot the sponsored link if you spend more than half a second on it before clicking. A yellow background and text that says 'Sponsored Link' is way less deceiving than the lack of a 'like' button.

The devil is in the details. And the significant amount of justification you do in your post shows that you're trying to convince yourself as much as us that it's OK.

Re: Image Ad Blending Works Really, Really Well

#195
post #133

Earlier quoted context omitted.

But with advertising, the attention transfer is involuntary . The user is always trying to spend their attention on something else, and they aren't getting what they pay for. If attention is currency then advertising is clearly bait-and-switch fraud.

The user is exchanging some portion of their attention to receive whatever service the publisher is providing. If you want to watch this football game, you give some portion of your attention to our advertisers. If you want to view whatever the regular content is on the site patio11 advertised on, you give some of your attention to patio11. There's nothing involuntary about it. If you don't like ads, don't watch TV s…

But we don't agree in advance to give up our attention in exchange for things. Nor do advertisers feel obliged, per se, to give us something in return for looking at ads. It is simply assumed that we don't own our attention, and it is up for grabs to anyone who can get their hands on it.

We expect to see ads in many situations, but only due to past experience, not because we ever made a deal with anybody.

And the most effective ads, the ones that advertisers strive to create, are those that are the most unexpected.

Re: Image Ad Blending Works Really, Really Well

#197
post #29

This reminds me of those newspaper ads that are crafted to look exactly like a newspaper article (columns of copy, headline, etc) but only set out with an " Advertisement " in very small print at the top of their enclosing box. I usually can pick them out because the font isn't a perfect match, but once in a while I don't notice until like a third of the way through the "article" when I'm going, "wtf? What was the ed…

The ad was explicitly approved by the editor. That's how BuySellAds works.

Precisely.

I sell adverts on my sites via https://www.buyads.com/ ( iSocket ), and I decline all adverts which are cheap and tacky, but approve adverts which are tastefully done and will appeal to the sector I'm in.

Point is: I'm in control.

Which is great, as I value my community and it would reduce their experience to take tacky adverts, but if it's something that they want and it integrates well, I'd definitely approve it.

The publisher is in control, Patrick gave the publisher the exact kind of thing that they want. Both parties benefit, as do the users. This IS the way to do advertising, where everyone in the equation is very happy with it.

Re: Image Ad Blending Works Really, Really Well

#198
post #62
post #40

Earlier quoted context omitted.

I'm curious as to why you'd draw a distinction between AdWords and scam filled advertising. Google's bread is buttered by credit cards, diet pills, and degrees. Search any relevant keyword if you don't believe me. Most of the advertisers I share pages with on the AdWords Content Network are also not pages you'd want your mother going to. The economics of Internet advertising heavily favor scams. Would it be better fo…

I don't know how to get any good data on this, but I think saying "Google's bread is buttered by ..." is at least accidentally misleading. Google supports a tremendously diverse ad ecosystem, some of which is questionable. However, at least in my experience, very few of my search queries involve such ads. The perception (if not provable fact) that they're mostly a class act is a big part of their enduring brand value…

I think when he was talking about ads for weight loss programmes etc he was talking about what comes up on the Content Network, not what is seen on Google's results pages. The Content Network is what you're in if you display Adsense adverts on your website.

Re: Image Ad Blending Works Really, Really Well

#199
post #162

Earlier quoted context omitted.

I never knew. How can that justify that they deceive many users?

Google's Adwords strategy was a reaction to competitors who literally sold placement in search results .

No. Google's initial reaction (long, long ago) to competitors who literally sold placement in SERPs was to not do that, and not have ads at all, and promote this widely.

Later when they got ads and "sponsored results", which they had to do of course because at some point you've got to make money somehow, they were very very careful to distance their ads from search results, because after all they'd just been promoting themselves as "not selling their search results" and of course "Don't be Evil" (which they dropped by now right).

And ever since, their ads have been creeping closer and more visually similar to their search results, right up to the point where, as the OP mentions, some of these ads look exactly like search results, except highlighted. Sure they are marked as "sponsored result" in very tiny light-grey letters but so were some of their search-result selling competitors back in the day.

So yeah no sorry, I fully understand this is the way corporations inevitably go as they get big and successful, but as far as I see it, Google has been distancing itself from this "evil" behaviour at first, and then, after competition was crippled, slowly crept back to--well not the worst examples of this behaviour, but it's still bad enough that if Google had been selling "sponsored search results" in this fashion right from the start, nobody would have bought their bullshit about "we don't sell our search results".

What they're currently doing is only marginally better than silently hiding sponsored links within the search results, as from every approach they're going through lengths to make this ad look exactly like a search result, except they highlight it and tiny letters say it's an ad (the 8px kind that UX blogs will tell you a lot of people have trouble reading).

I even just noticed they gave their "top search result ad" exactly the same ">> Preview" button on mouseover as any other search result! Sorry but there's no question that what Google is technically doing right now, is selling their top number 1 highlighted "search result", misleading people to look exactly like a search result, except in a token-attempt at "honesty" it is marked as an ad, except you know as well as I do that the less computer-savvy people don't even notice that, and just click it if it seems to answer their question, without realizing it is an ad. Just as if it was a search result.

I mean, hey I think it's bullshit, sure they're allowed to do this, but let's not pretend Google is not selling (top) placement in their search results, and that this top ad is not desperately trying to look and behave exactly like a search result in a deliberately misleading way. They are no longer any better than the competitors whose ad model they were so against back in their early days.

Re: Image Ad Blending Works Really, Really Well

#200
post #52
post #49

Earlier quoted context omitted.

Can we just note again here that in your construction, "the bad guys" include Google themselves? The tactic we're talking about is, as Patrick notes at the very start of his post, the same one Google uses at the top of every SERP.

No. Only if Google included barely distinguishable paid placements in their organic results would it be a fair comparison. I'll concede that some minority of users might not be able to distinguish paid results from organic on Google, but no one could do it with these "blended" ads -- that's why he's getting such a high CTR. Users are being tricked.

> No. Only if Google included barely distinguishable paid placements in their organic results would it be a fair comparison.

Well that is what they are.

The only distinction is the 8px letters that say "Why this ad?" which are hard to read and widely ignored.

> I'll concede that some minority of users might not be able to distinguish paid results from organic on Google, but no one could do it with these "blended" ads -- that's why he's getting such a high CTR. Users are being tricked.

Let's call it the majority of users except for programmers hackers and other computer-savvy folk, and I'd say you got it about right.

Have you ever seen a kid use Google? An elderly person? Your parents? Shoulder-surf a random non-coder in a coffeeshop or a library? These are the kind of people that type "google" into the browser search box to go to google. I work with 8-12y kids that come specifically to learn about websites and computers, and among those already savvy interested kids there's only about half of them that understands what the difference is between a paid ad on Google and a search result. They'll just click it if it appears to lead them to whatever they were looking for.

I agree that these ads are "blended" way more than Google's paid highlighted top search results/ads, more deceiving, but it really is the same thing. The "blended" ad also clearly mentions in a very tiny font that it's a sponsored result.

I would also think the only way that we're able to more easily recognize the Google paid highlighted top search result/ad is because we use Google every day, we expect it's there, and from previous usage we know that search results that look a bit different are probably sponsored. The "busyteacher" website is only different in the sense that its users have no known expectations about the site because they only use it rarely.

Post reply on HN