This is hilariously accurate.

I've been predicting for a while now that sites would fall back to the old television show model "RockAuto presents MustangForums.com" or something to that effect.

Instead, we get a dancing Albert Einstein begging us to take IQ tests.

This is in fact very popular among both YouTube channels and podcast producers. It makes a lot of sense for an advertiser to pick a producer who matches their market and have that producer create an in content advertisement. I am fine with this. The content producer wins, the advertiser wins, and the viewer wins by getting better ads by someone they want to support.

One big problem with entirely static ads is that websites are global but the ads running on them are for brands that are likely local (at least to a specific country.) If I visit the NYT website in e.g. Norway, should I still see ads for an American brand of paper towel that doesn’t exist here; or should I see ads for Norway paper towel brands?

The flip-side of this is that I’ve noticed that YouTube shows me PSAs from my own municipal government (“there’s an election soon” ads, “we’re building a new piece of civil infrastructure” ads, etc.) I actually kind of like that; I don’t have cable, so it’s not like I would see them anywhere else.

The entirely-static ads model does work when the consumption of the media is entwined with the consumption of the advertised brands, though. For example, a podcast can certainly advertise its own tour, since—given that you’re listening to the podcast—you likely want to see the podcaster speak in person, even if you can’t make it there.

Or, of course, if a (global) website is just advertising another (global) website. The NYT can advertise Amazon just fine.

> If I visit the NYT website in e.g. Norway, should I still see ads for an American brand of paper towel that doesn’t exist here; or should I see ads for Norway paper towel brands?

That's still possible with static ads. The server can simply lookup your country from your IP address and serve the relevant ad, without tracking you at all.

Geotargeting can be done on the server-side without involving any trackers, 3rd-party javascript and what-not.

This is a good question that I've thought about a lot because the services who advertise with us are regional.

We fully control and host our static ads and try to keep them high quality, so I've decided that minimally using IP to loosely serve a more relevant ad is okay. MaxMind offers a downloadable Geo IP database that we use to do this and are not needing a 3rd party service for this.

This not necessarily a problem. There's nothing to stop the website operator calling out to an ad provider, with the ip/location of the user, and getting an ad to embed.

The upside for the user is that location and whatever the one site is able to determine about the user is all that can be shared. If the user hasn't logged in with their real name - that probably isn't much.

The issue is that you can make considerably more money using ads that 'track' you.

So instead of one ad being enough to pay for your content, you have to fill your website with banner ads, embedded ads, scroll over ads, animated ads, etc etc.

It's a slippery slope, more people use adblockers causing content creators to add more advertisements to generate the same amount of income. More people are bothered by the increase in ads, and download adblockers themselves. Rinse and repeat until ad supported content is unrealistic for all but the biggest of websites.

And I'm pretty sure even checking location is controversial. I've at least seen it included as part of tracking in the past.

> (“there’s an election soon” ads, “we’re building a new piece of civil infrastructure” ads, etc.) I actually kind of like that; I don’t have cable, so it’s not like I would see them anywhere else.

I don't have cable, either, but I do have a pair of rabbit ears to keep up with local news via OTA broadcasts.

Ads can still be geo-customised entirely from the server side, although that's a bit more work than just throwing an image in a directory.

This is what we do at Office Snapshots (https://officesnapshots.com).

We post about office design and our ads are primarily for office furniture or other services related to the industry. We also self-host the ads which are non-animated jpgs and sell them without using any ad networks.

What you describe works well for us :)

We started doing that too, in addition to Google ads. We deal directly with the advertiser, self host a specific ad for them. It's MUCH more profitable than Google ads. Will phase out the Google Ads soon.

Minimal work involved. A bunch of emails back and forth, where we tell them our ad size if they provide an image. Or they provide a logo and we manually ad the text if they aren't technical enough to provide a custom image.

What type of overhead is associated with this? Are you experiencing an increase in server costs? What about the resources it takes to build those relationships, maintain the content, or manage the infrastructure?

I'm very interested in learning more...

We use WordPress and plugins which offer these features so hosting and server costs are pretty minimal.

It definitely takes time and work to develop and maintain relationships, but we also get to keep 100% of the revenue. In some ways we've just decided that ownership of the relationship and process is more important than being able to quickly slap some Adsense code up on the site.

That said, we also used Adsense early on, but have been doing this for ~5 years.

I'm also wondering how revenue compares. Every one clearly seems to think that "perosnalized" ads are worth more; can you offer any insight into if you make more or less with personalized vs your approach?

People seem to have forgotten that users are most likely to be interested in what they are currently reading, because it is what their mind is currently focused on.

You can get a sense of advertising fees by looking at comparable newspaper ads and industry trade magazines. Even take a look at billboard or radio ads. Custom ads will be an order of magnitude higher that what you get from Google.

I wish I had a deep and thoughtful answer, but a simplistic one is that we do get to keep 100% of the revenue for the ads we sell so in that sense we make more.

Because our content is so specific, ads which are relevant to the content end up targeting the user because you wouldn't be spending time on the site unless you care about the content.

So how do you know how much to charge? Do you relate it to AdSense's cost plus Google's profit or you follow some other practice?

Initially I just set a price I thought was fair and have adjusted accordingly as we've grown. I do recall that the amount we received from Adsense was less than what we ended up choosing to charge for the same space.

We also sell the space per month as opposed to based on impressions or clicks so it makes it a little more straightforward.

> What's wrong with the "old" way

A couple things:

1. It facilitates a world in which only large content providers, who can afford to individually sell ads to advertisers, to exist.

There's a lot of overhead to ad sales and individual companies do not want to work with 1,000,000 providers, they want to work with 10-100.

2. It's substantially less efficient and only works for brand advertising or mass-market direct-response advertising.

One of the greatest things that Facebook and ad retargeting enabled was the rise of direct-response brands. Previously if you were selling a niche product - and most larger brands started out with a niche product - it was very difficult to reach an early audience who would be interested in purchasing your product. Facebook and Google flipped this on it's head, enabling millions of businesses to more efficiently reach customers. Facebook alone made the direct-to-consumer brand explosion we've seen over the last 10 years possible.

Tracking ads are collecting data on people without their explicit consent or knowledge. The trade in bulk personal data is a stain on the internet.

I am quite content to block these sorts of adverts and I'm not worried by the site's loss of revenue. I am not responsible for their slimy choice in business model.

Let's get back to context based adverts like DuckDuckGo use. There was no need for the internet to take this path - it only did so to rapidly monetise after the dot-com bust blew their VC funded rapid growth plans out of the water.

> There was no need for the internet to take this path - it only did so to rapidly monetise after the dot-com bust blew their VC funded rapid growth plans out of the water.

No need, aside from the trillions of dollars of economic value it created and hundreds of thousands of previously impossible businesses it created.