3 ms·

It's relatively easy to debunk/prove your point about targeting. Test both on the website you control. At the very least you can geolocate the visitor and prese

It's relatively easy to debunk/prove your point about targeting. Test both on the website you control. At the very least you can geolocate the visitor and present an ad in the most appropriate language about a product in their local store. I'm pretty sure targeting would win by a wide margin. Google ad guys are the best in business, and you're basically implying they are in a weird targeting cargo cult.

ggm 9y ago
No, I think the entire ad economy is a weird targeting cargo cult. I don't think they're any worse than anyone else in the space. Google ads: serving me pictures of the shorts I bought three months ago, on a daily basis. Seriously? You say they're the best in the business, and their business is selling to an advertisor "he bought a pair of cargo shorts three months ago" as a reason to sell me .. cargo shorts.
- tlack 9y ago
  I'm with you here. I was having an argument with someone about how Amazon has already conquered the target-upsell-intelligence space, and just as he was saying that, I saw numerous ads for something I already bought - same pictures, different sellers. How many of this lens do they think I need? This is a great example of something being an obvious correlation in "computer science land" often doesn't really bump the numbers as it might.
- morgante 9y ago
  > You say they're the best in the business, and their business is selling to an advertisor "he bought a pair of cargo shorts three months ago" as a reason to sell me .. cargo shorts. Someone who bought cargo shorts 3 months ago is far more likely than a random individual to buy cargo shorts tomorrow. You're thinking about this from the perspective of a single individual consumer, but that's not how the ads business works. Ads are bought in the form of aggregate possibilities. If I were a cargo short vendor, I'd happily pay considerable sums to reach everyone who had recently bought cargo shorts. Maybe they're unhappy with their cargo shorts and want a different brand. Maybe their shorts tragically broke after bending over too far a few weeks after purchasing. Maybe you loved them so much that now you want a separate pair for every day of the week. All of these are reasons that betting you'll buy cargo shorts is a much better bet than betting on a random person. People buying ads are not idiots. The ones buying digital ads are the smartest ones and many have strong backgrounds in analytical fields. It's arrogant for you to assume otherwise just because you're sick of cargo short ads.
  - ggm 9y ago
    well.. arrogant is a strong word. anyway, your general case is that if you express any interest in some thing in time, then you are better-than-random to target. I'd posit that its a spread, and whilst its an edge, its nothing like 100% foolproof. I also buy ads (I help run experiments using google ads to run HTML5 in the browser to measure IPv6) and I know how the ad bidding process works, and keyword matching and its pretty crude. Yes, it has age and other profile selection, but so does traditional print ad. digital ads are better than print ads because print is dying. I think the whole metrics thing, its pretty oversold. since it was known I visited cargo short buying and advertizing sites, its also known the exponential dropoff in not visiting them. their ageing out algorithm needs adjusting.

ggm 9y ago

No, I think the entire ad economy is a weird targeting cargo cult. I don't think they're any worse than anyone else in the space. Google ads: serving me pictures of the shorts I bought three months ago, on a daily basis. Seriously? You say they're the best in the business, and their business is selling to an advertisor "he bought a pair of cargo shorts three months ago" as a reason to sell me .. cargo shorts.

tlack 9y ago
I'm with you here. I was having an argument with someone about how Amazon has already conquered the target-upsell-intelligence space, and just as he was saying that, I saw numerous ads for something I already bought - same pictures, different sellers. How many of this lens do they think I need? This is a great example of something being an obvious correlation in "computer science land" often doesn't really bump the numbers as it might.
morgante 9y ago
> You say they're the best in the business, and their business is selling to an advertisor "he bought a pair of cargo shorts three months ago" as a reason to sell me .. cargo shorts. Someone who bought cargo shorts 3 months ago is far more likely than a random individual to buy cargo shorts tomorrow. You're thinking about this from the perspective of a single individual consumer, but that's not how the ads business works. Ads are bought in the form of aggregate possibilities. If I were a cargo short vendor, I'd happily pay considerable sums to reach everyone who had recently bought cargo shorts. Maybe they're unhappy with their cargo shorts and want a different brand. Maybe their shorts tragically broke after bending over too far a few weeks after purchasing. Maybe you loved them so much that now you want a separate pair for every day of the week. All of these are reasons that betting you'll buy cargo shorts is a much better bet than betting on a random person. People buying ads are not idiots. The ones buying digital ads are the smartest ones and many have strong backgrounds in analytical fields. It's arrogant for you to assume otherwise just because you're sick of cargo short ads.
- ggm 9y ago
  well.. arrogant is a strong word. anyway, your general case is that if you express any interest in some thing in time, then you are better-than-random to target. I'd posit that its a spread, and whilst its an edge, its nothing like 100% foolproof. I also buy ads (I help run experiments using google ads to run HTML5 in the browser to measure IPv6) and I know how the ad bidding process works, and keyword matching and its pretty crude. Yes, it has age and other profile selection, but so does traditional print ad. digital ads are better than print ads because print is dying. I think the whole metrics thing, its pretty oversold. since it was known I visited cargo short buying and advertizing sites, its also known the exponential dropoff in not visiting them. their ageing out algorithm needs adjusting.