4 ms·

Honestly this is well known among any decent marketer, but what the article seems to miss is the answer to the question they pose: >“Would this customer have b

Honestly this is well known among any decent marketer, but what the article seems to miss is the answer to the question they pose: >“Would this customer have bought without this ad?” Even if the most likely answer would be "Yes", I think the proper question would be: "Would this customer have bought MY PRODUCT without this ad? Here the question is most likely "No" (unless we're talking about retargeting but let's leave that out). Who would they have bought from if they haven't seen your ad? They would have bought from someone else. So, even if personalized ads are pointless, these companies are right about their advertising offers being a driver for sales. They could drop the whole "personalized ads" angle and they'd still be right since they would keep driving results. Now are all the results sustainable? Down the line, they won't, because the more players get in the bidding game the higher the cost per results will be - and no personalization will help small businesses on that front.

perl4ever 5y ago
If you want money from advertisers with the least effort, don't you have an incentive to identify the people most likely to buy before you advertise to them? And thus do the least possible amount of converting people to products they weren't going to buy? It seems to me that you're hypothesizing, say, Google, doing more effort than necessary for no reason. Regardless of how much ability they do or don't have to affect behavior, if they can sell correlations, worrying about causality is wasted effort. Causality determines if they are cheating the customer but not if the customer is paying them, so ignorance is bliss. The way I read your comment is that we should believe what's inside a black box is a valuable product because it hasn't been proved not to be. I'd argue the label on the box is zero evidence of anything, and the incentives not to look inside are evidence that nobody is putting good stuff in it.
- libertine 5y ago
  >If you want money from advertisers with the least effort, don't you have an incentive to identify the people most likely to buy before you advertise to them? And thus do the least possible amount of converting people to products they weren't going to buy? Yes, but this is what they do, that's why I said "any decent marketer knows this". Facebook, Google, have different "buckets" of audiences based on their behaviors, and they identify those who are more prone to click, to engage, to buy. And we, as advertisers, can choose who we want to reach. But what I got from the article is that advertisers are just throwing money away because those consumers would buy anyway. >It seems to me that you're hypothesizing, say, Google, doing more effort than necessary for no reason. I didn't understand what you're saying here. If by effort you mean new features, then well... someone has to sell this upstream to justify marketing budgets. Shiny new things help convincing people. >The way I read your comment is that we should believe what's inside a black box is a valuable product because it hasn't been proved not to be. I'd argue the label on the box is zero evidence of anything, and the incentives not to look inside are evidence that nobody is putting good stuff in it. You have no other way but to believe it when you commit your budget, and the results show. If you close the faucet, sales tank. So for sure everyone knows it works. But of course we're talking about humans, and we're not conditioned machines that just react to external queues... so it's not just show up and they'll give you money.

perl4ever 5y ago

If you want money from advertisers with the least effort, don't you have an incentive to identify the people most likely to buy before you advertise to them? And thus do the least possible amount of converting people to products they weren't going to buy? It seems to me that you're hypothesizing, say, Google, doing more effort than necessary for no reason. Regardless of how much ability they do or don't have to affect behavior, if they can sell correlations, worrying about causality is wasted effort. Causality determines if they are cheating the customer but not if the customer is paying them, so ignorance is bliss. The way I read your comment is that we should believe what's inside a black box is a valuable product because it hasn't been proved not to be. I'd argue the label on the box is zero evidence of anything, and the incentives not to look inside are evidence that nobody is putting good stuff in it.

libertine 5y ago
>If you want money from advertisers with the least effort, don't you have an incentive to identify the people most likely to buy before you advertise to them? And thus do the least possible amount of converting people to products they weren't going to buy? Yes, but this is what they do, that's why I said "any decent marketer knows this". Facebook, Google, have different "buckets" of audiences based on their behaviors, and they identify those who are more prone to click, to engage, to buy. And we, as advertisers, can choose who we want to reach. But what I got from the article is that advertisers are just throwing money away because those consumers would buy anyway. >It seems to me that you're hypothesizing, say, Google, doing more effort than necessary for no reason. I didn't understand what you're saying here. If by effort you mean new features, then well... someone has to sell this upstream to justify marketing budgets. Shiny new things help convincing people. >The way I read your comment is that we should believe what's inside a black box is a valuable product because it hasn't been proved not to be. I'd argue the label on the box is zero evidence of anything, and the incentives not to look inside are evidence that nobody is putting good stuff in it. You have no other way but to believe it when you commit your budget, and the results show. If you close the faucet, sales tank. So for sure everyone knows it works. But of course we're talking about humans, and we're not conditioned machines that just react to external queues... so it's not just show up and they'll give you money.