When you're browsing around looking at office design projects and photos, you're seeing static advertisements for office furniture and services you might be interested in if you're in the industry. And if you're spending time on a website that consists only of office photos, you are probably are in the group interested in the advertisements.
I would be interested to know whether you have any stats to back this up; if it works this would be great.
To actually judge whether users wanted to see the advert the best metric would be conversions, but that's likely data the site owner wouldn't have (and even with that data, you'd need something to judge it against - for all the talk that most adverts are ignored, the fact remains that some adverts through Google/Facebook/whatever ad-network can lead to direct conversions with a positive ROI).
Not saying it wouldn't be interesting to hear what his typical CTRs are, just that nobody should look at the answer and think you can make significant judgements based on them.
Setting aside what's "average" - are you controlling for the type of advert (both what it's advertising and its format), the type of audience, etc...
Over the years I've seen plenty of examples where CTR and ROI just aren't linked at all. Right now my work doesn't involve a huge amount of this kind of advertising (though I do have experience overseeing 7 figure budgets in the past) so I don't have much great data to pull examples from, but here's one real example from a couple of months ago:
Two not-huge (sub $10k) media buys, both direct with websites (not through networks), both with websites that are well-respected by their readers and by their industry, both using the same adverts. One of the sites excitedly sent us a report about how well the ads had performed, a 5.63% CTR for banner ads, clearly we're popular! The tracking on our side showed that of the just over 15k clicks they sent, 12 converted into users for us. Meanwhile the other site was reporting a 1.32% CTR, sent just over 6k clicks (pricing and order size wasn't exactly identical between the two sites), and landed 700 new users. I know both companies well enough to not suspect click fraud, and I've seen the company that performed badly perform well on other campaigns. But if you just looked at this comparison, there's no argument that the campaign with a much lower CTR was actually hitting an audience that was more interested in what we had to sell than the one with the high CTR. There's lots of possible reasons, from accidental clicks depending on how sites are set up to the audience targeting each site did (both are big enough sites to offer specific targeting within their audience), etc.
(Edit to add that of the two campaigns I compared, both produced a negative ROI from my point of view, i.e. our CPA was far too high - just one was a lot worse than the other. I just pulled up the first spreadsheet I saw from the most recent obvious example I could think of for CTR != ROI.)
If CTR is the only metric you are able to look at, then you can do your best to make judgements on it. But the reason the adtracking solutions that annoys so many people exist is because without proper tracking, it's really hard to actually judge the value of any advertising you buy.
All that said, CTR does still have a big place in the industry, because generally speaking websites (or ad networks) are able to tell potential customers their typical CTRs, but aren't able to give data about further down the funnel (conversions, etc.) because this is data the ad buyers control, not the ad sellers. But it's really important not to think it tells the full story.
Also, by ~5% CTR in your example are you meaning that for every 100 impressions that campaign was getting 5 clicks?
Depending on the ad format/placement and many other variables, it's possible to have a high-ROI advert with a 0.1% CTR, or a low-ROI advert with 5% CTR (or the exact opposite).
The majority of companies which are the best candidates for advertising on Office Snapshots do not have e-commerce stores to point users to as most of the sales in this particular vertical happen via office furniture dealerships.
There are some exceptions, and I've tried some affiliate links in the past, but the types of people I'm marketing at are people who will at some point in the future purchase 100 chairs for a project they are working on and not people who see an ad for a chair and want to buy it immediately.
I've felt for a while that I liked facebook when it was weirder and there was more diversity in the types of content, but that has really declined in their pursuit for optimizing engagement. Twitter still has this so I appreciate that network quite a bit more for it.
If that is true, the problem will solve itself eventually.
As someone with more experience than most advertisers when it comes to native ads (i.e. outbrain, taboola, revcontent, etc...) I can assure you that clicks and impression costs are going up, not down. As people learn what works and the competition gets fiercer, the costs go up.
The race to the bottom you fear is actually, a race to the bottom of what advertisers can afford to continue advertising. Right now, Native ads are dominated by supplements, which is essentially snake oil salesmen, and also some casino, and other either mass appeal or shady enterprises. Most legit businesses are priced out of the game...Paid Search and Paid Social have done a good job keeping out these low quality advertisers...
Essentially, the race to the bottom of creating clickbait that you are pointing out, is actually a race to the top. A race to the highest margin, most mass appeal advertisers.
If you really think about publisher incentives, they are more likely to create clickbait when the value of clicks goes up, not goes down. At some point of time when prices drop, it becomes easier to make money in other areas...so why bother.
Edit: corrected spelling on a word
I believed in the promise of the Internet to create a more informed, compassionate, open, and enlightened society and that hope is turning into disillusionment.
This is problematic for me. The more a site knows about me, the better and more interested are the ads. But at the same time, it gets more scary.
Look for an item on Amazon, and you will see it advertised in a lot of pages that you visit. It can be useful to remember that it is there, also tempting. But it is the same feeling that being at home and when you look back there is the guy from the gadgets shop with the camera that you were looking in his hand. Creepy. Scary.
Somewhere along the line (probably with the demise of AOL) it was decided that the model of the internet wasn't going to be such that we paid a subscription for the content we consume, but that the model would be akin to freesheets and completely ad funded. It's natural that people who invest in that kind of advertising are going to want some kind of metrics in return, and probably also natural that publishers will try to optimize for that metric too.
Im taking a genomics class, so I figured I'd look up to see if there are any videos for more information on some topics...
Turns out there seems to be a cottage industry of having the computer read a definition, and post it on YouTube: for example this user. These videos show up often in searches and crowd out some of the more interesting stuff.
Wastage
And as more and more ads are viewed and not clicked, (there's probably some adtech term for this) the price per view is going to go down, which will encourage more clickbait, which will drive the value of an impression down, etc.
The article struck me as a bit weird, it complains about yellow journalism, but it doesn't really dig into why it's happening. It just declares that shitty articles are written for the metrics, and leaves it at that. As if "metrics" had any intrinsic value. The mapping of metrics to money, as done by internet advertisements, is the actual problem.
While engagement is certainly a way to get more people to view more ads, it's but a tool and symptom of the system it's a part of, not the genesis of it.