Problems with self hosting:
- It's super duper hard to change the status quo. For them, the current solution 'works fine', once they fight the supposed evil that ad blockers are. Trust me, I fight for this sort of stuff and they just don't understand things the way we do.
- You can't do as much or sell ads for as much. You can charge more if you know more data about your visitors and can sell the ads to clients under the impression that it'll reach the exact demographic they're looking for.
- For some companies, it's already too much of a chore to have an in-house dev team to manage the content website. Have to have _people_ manage all these self hosted ads is not appealing to these companies. Very similar to AWS vs self-hosting
- An image is less interactive, so you can't sell it for as much. Those big fancy ads that are interactive are sold more many many more times the price as just a single jpg more.
- If you're 'lucky' to work in a media org big enough to have a sales team, even they don't sell out every single ad placement all the time - there's a big tailed on these states they still want to monetise and ad networks helps them do that.
That aside, things are changing. Everything on our site happens over HTTPS, which is a plus. We do win 'arguments' to move things in-house and not served from an ads network, where appropriate. We have an in-house 'studio team' who creates custom campaigns and creatives for clients, and they're almost entirely self hosted.
I'd just like to add that almost all the problems with malicious ads are caused by ad networks that are used as "backfills" for when the in-house ads aren't booked. Backfilled ads are not inspected and can contain everything, but they are also needed because otherwise you'll have unsold inventory.
When we don't have a big sponsor for a section of our site, or haven't had enough direct sales for certain placements, it goes out to the ad exchange to action the spot off. That's where the problem is. That's somewhere we could start enforcing certain quality.
Ad exchanges and real-time bidding (RTB) are only causing a problem for people working in the media space because on the one hand if you don't have them you lose money, but on the other hand it's like gambling in regards to the quality of ads that will be served.
If I was buying a box of cereal, "we couldn't source enough reputable corn, so we topped it with melamine" would be the death of that cereal, for me.
(And yes, this is buyers at the cereal company instead of ad sellers, but it's the same: the hypothetical cereal company was not able to "sell" enough of its money to corn suppliers.)
Of course the goal is to operate at 100% efficiency, but in reality that's not going to happen all the time. Backfill channels like this are a way to make at least a little bit of money when sales a slow for a bit for whatever reason.
In terms of the problems the online media industry faces with advertising, the efficiency of in-house sales teams is not a huge priority atm.
If part of your pitch is that you avoid adblockers, then once you get big enough you'll constantly be fighting an uphill battle.
Obviously in execution it would be different, but I hope you get the idea.
Edit: I think self-served ad platforms are one way forward, but it depends on what problem you're trying to solve.
[1] https://github.com/gorhill/uBlock
* Either a /static/ image or colored text
* Normalized (re-packed, stripped of all other data, etc)
* Vetted by a human somewhere to not be 'misleading in appearance' (not checking facts, but making sure that it doesn't look like a download button or anything else not obviously an ad)
* Based on the content of the context and nothing else (you can bake them statically based on the story and/or comments thread on the page)
I read it as a measure of their financial desperation. They're eating their seed corn (their reputation is their most valuable asset in that without that, what else do they have?), and, well, what other choice do they have? HN has been talking about the other choices for years, but every time I see someone try one, they sink.
I don't think shifting responsibility to publishers is the answer. A half-serious idea would be to have a neutral regulatory body issue "licenses to advertise" but how long until corporate interests took that over?
At the end, I know they need to get money but ads don't serve any purpose. I mean I never buy product from ad, I have only things that suite what I have search 10 minute ago. Why putting ads for that? I already know the product and this will never push me to buy something.
Really, they're mostly spinning a web of lies that keeps them employed. You won't find employment doing this for a living if you don't pretend like most of what you do really impacts everyone.
I'm one of those people who will actively abandon products that market too heavily or with material that manipulates its consumers. I switched away from Dove soaps/shampoos back when they started their viral "Campaign for Real Beauty" shenanigans. Pretending you care about your customers' self-image problems... to market soap... turned me off.
It's the difference between Coke & Budweiser, and Squarespace & Geico ads.
There are only so many ways to obfuscate your DOM without screwing up the page layout and making your UI people angry. The battle against add blockers cannot be won.
I strongly believe that the ad industry will die, at least in the sense of what it is at the moment. Fighting ad blockers is its death rattle. That doesn't stop the technology being interesting though. Quite the opposite in fact, as a developer I find it fascinating.
There is no reason a neutral 3rd party can not review logs, software, or other data to ensure in house ad metrics are accurate.
(I work on the ad systems team of an affiliate marketing company, where essentially all advertisement contracts are done on a PPA basis, rather than PPC)
Because the content of the site is narrow (photo tours of office design projects) it ends up attracting a certain group of people who are interested in office design and office furniture (self-targeting if you will).
Any moderately popular news site will have filter authors going out of their way to block their ads in the filters regardless of URL domain.
I feel like there's this myth that adblock is just trying to block "bad" ads, but the evidence shows most blockers really try very hard to block all ads. Trying to stay one step ahead of that is a losing proposition for media companies.
Well, I'm sure that it would not be impossible:
- ads are the changing part of a webpage, so they can be identified
- ads could be collaboratively removed, by allowing the user to select the ad, then send its fingerprint to a central entity, so that others can benefit
- AI techniques could be used to identify and remove ads.
That is kind of the point. Online ads should go back being exactly like print ads are. Sold individually, static, untargeted and un-trackable. Not because the ad-networks, advertisers or site owners think its best, but because its the only type of ad that the viewers will accept eventually.
And maybe I should clarify, how does self-hosted ads solve the problem?
- It's hard to block ads on the same domain as the main content.
- Just loading an image with an ad is not as much of an performance hit as javascript loaded tracking ads and thus should not annoy users as much.
- More secure, because you can not hack one single ad service and distribute mallware to thousands of sites. Besides, the solution is less complex, and complexity always leads to bad security.