13 ms·

I’ve always thought that the “advanced tracking” touted by some in AdTech was: 1. Generally not that advanced (yes I searched for hotels in Florida and so yes

I’ve always thought that the “advanced tracking” touted by some in AdTech was: 1. Generally not that advanced (yes I searched for hotels in Florida and so yes I’m probably taking a trip to Florida... don’t need fancy ML to figure that out.) 2. Not that smart. (Hey we actually got back from our Florida vacation two months ago... don’t need to keep seeing hotel offers for Florida!) It totally makes sense that just “dumb” targeting (show people adverts related to the content on the page they are currently looking at, which requires no tracking) might actually work better than all the black magic products companies promote.

afiori 6y ago
The best argument for the usefulness/profitability of deep tracking is simple statistical inference. As in if there are 10000 people that with browsing patterns similar to yours and non-trivial fraction of them one day show interest in a discount/service/trip then you are also likely to be interested. It is the official explanation of how facebook/google can magically listen to conversation IRL that they were not supposed to hear: we simply are not as unique/unpredictable as we would like to think.
- baybal2 6y ago
  I think it's one of the best argument against. It was at least 5 years ago when I started to notice businesses "A/B testing themselves to death" — "It makes no common sense, but increases sales," and then it does until it doesn't, and you are left with no idea what your client actually wants, and no idea how to get out of that "A/B tested to absurdity" situation
  - mumblemumble 6y ago
    This also seems to be an inherent danger with thinking that data science == machine learning == AI. I don't want to fundamentally challenge the ŷ-centric approach approach of machine learning. But I do want to say that there's a lot of danger in taking it too far, and thinking that β̂ doesn't matter as long as ŷ is good enough. β̂ has a habit of not mattering until, very suddenly, it really matters quite a lot. And it's even more dangerous to think that an algorithm for generating formulas is somehow doing something intelligent just because you like what you're seeing in the ŷ department.
    - afiori 6y ago
      As someone that is not "in" with those fields what do ŷ and β̂ stand for here?
      - mumblemumble 6y ago
        ŷ is the prediction you get out of the formula, β̂ is the coefficient(s) on the variables in the formula. So, take the simplest possible model: You think that your dependent variable is linear in the dependent variable, and that the line passes through (0,0). That gives us a simplified version of the classic equation for a line, where we say that y is equal to x times some constant that indicates the slope of the line: y = mx In machine learning (or statistical inference), the variable conventions are slightly different, so it would look more like: y = βx And then, since this is machine learning (or statistical inference, or whatever), we don't know the true equation, we're just working with an estimate we've inferred. We stick hats over the estimated values to make it clear that they're estimates: ŷ = β̂x What I was describing, then, is which side of the formula you're focusing on. Traditionally, statistics has been all about scientific inference, and trying to figure out and characterize causal relationships. So you want to know "If I change x in some way, how will that affect the outcome, all else being equal?" β̂ is your tool for teasing those things apart. Machine learning is typically cast as being more about making predictions from observational data. ŷ is just the formal notation for describing those predictions. It's pretty common in machine learning to say, "We are working with observational data, we're not worried about causality, so we'll only pay attention to the predictions." There can be value to that approach, but only to the extent that you really don't care about causality. Which. . . surprisingly often, β̂ matters, even when the people who built the model were trying to ignore it. It becomes especially important when you're using the model to make decisions. For example, when a company is using a machine learning model to try and identify candidates for a position they're trying to fill, they should absolutely be paying attention to what the model thinks about people who liked Lane Bryant on Facebook, and how that compares to people who liked Jos A Bank, and whether they really want things like that to be playing into how they make hiring decisions.
  - cortesoft 6y ago
    Yeah, I often wonder this... a/b tests only ever measure short term differences. What if the change you are making increased sales for a few months, but pisses off users in the long run? An A/B test isn't going to show that.
    - shostack 6y ago
      A regression model might if you have sufficient data.
      - cortesoft 6y ago
        You would have to run the a/b test for quite a while, and I don't think many companies do.
- code4tee 6y ago
  I see that but I think the broader point (and the point of the article) is that too often “success” in finding some less obvious link actually comes at the cost of overlooking the simple stuff. In the end there’s an argument that while neat the fancy ad targeting actually fails to meet the core business objective of maximizing ad performance.
  - afiori 6y ago
    My personal opinion is that for most companies having an organically grown audience is more valuable than a lot of faceless traffic. Obviously they can help each other if you do things right and are lucky.
lowwave 6y ago
It totally makes sense that just “dumb” targeting (show people adverts related to the content on the page they are currently looking at, which requires no tracking) might actually work better than all the black magic products companies promote. This is exactly kind of "if all you have is a hammer, everything looks like a nail" that turns me off about AI/ML community.
- rhizome 6y ago
  Just add 300 DNS queries and wait for the RTB to complete!
Swizec 6y ago
My girlfriend is currently experiencing the dark side of smart tracking. Female who hit 30 and is in a stable relationship. The ML just can’t figure her out. One day every ad is for dating sites, the next day it’s all babies and pregnancy, then engagement rings, wedding locations, more dating sites, a few days of houses to buy ... it’s like the algorithms can’t decide what life stage she’s in.
- ceejayoz 6y ago
  > it’s like the algorithms can’t decide what life stage she’s in Or, it's working on figuring that out.
  - synthc 6y ago
    When my first kid was born, all the ads I got where either baby clothes, daipers etc., or Second Love. A clear separation between A and B
    - mcv 6y ago
      Makes me wonder if there are a lot of people who start cheating once they get a child. What exactly is that algorithm trained on?
  - phkahler 6y ago
    Just like she is right? Are you going to marry this guy or move on? Are kids in your future? How about a house, are you two moving in together? Probably all the same crap nosey family and friends want to know. Are advertisers an omnipresent nosey entity? Probably :-)
- harry8 6y ago
  That's what you want. You don't want these assholes knowing your business. The classic example is when they start advertising a whole lot of pregancy stuff to someone who doesn't want to be pregant and for whom that might be a serious personal problem - advertising seen by others who she does not want to see that. Dumb algos means freedom and privacy. Smart algos that actually work stuff out are dangerous and vile. Nobody gave them permission for that. Nobody complains about seeing ads not specifically tailored for them.
harikb 6y ago
This kind of page-content-based ad targeting did exist way before fine tuned retargeting (say, search or audience/segment retargeting). There are whole bunch of companies that serve this exact market. The reason a more complicated retargeting was created is proof that there is value in it, for advertisers. I am not claiming this is good for people - I am talking only about the business side of things. > don’t need to keep seeing hotel offers for Florida This is not for the lack of smartness in code - it is that the conversion funnel is so narrow that "saving that $30 per month, by not showing you the ad, on a $1m campaign" is not worth tracking and excluding. Although I do believe companies should take the extra effort to avoid the annoyance and be a good citizen - particularly once they have identified "the one real person who actually bought something" among 99 other bots/crawlers/untraceable entities. Not all ad companies appropriately value the real person and the annoyance caused to them. Some do.
- mumblemumble 6y ago
  Ad companies being able to figure out that I just reserved a hotel room, so they can stop advertising hotels to me, might reduce the annoyance level. But I'm also pretty sure that, if I found out that they had done this, I would never patronize that hotel again.
  - harikb 6y ago
    I am talking about cases where the ad results in a click or view through and finally tracked to a purchase from the same provider. You shouldn’t see the same ad again or another ad to rent a hotel room from same hotel chain. I thought that was the case GP was taking about. I am excluding the likes of companies like BlueKai (now Oracle) who facilitate trading the cookie to other companies for further targeting or completely different chain of ads based on a search click to another hotel chain That said, all of this is getting killed by recent Safari and Chrome change of getting rid of all 3rd party cookies
- valgaze 6y ago
  @harikb, what about something like this: https://i.imgur.com/LYU6NBb.png https://i.imgur.com/LYU6NBb.png Surrounded by content on advice "blue light"-- gets targeted for Blue-tooth lightbulb
Traubenfuchs 6y ago
Amazon showing me a selection of typical buy-once products of the kind I just bought is a depressing sight. They have infinite money and his is the best they can come up with?
- grandmczeb 6y ago
  I don’t understand why this is such an obviously bad strategy - lots of people buy an item, cancel or return it, and then buy a similar item.
  - cgriswald 6y ago
    Let’s say you buy a washing machine on Amazon. You’re not likely to need more than one. Amazon knows whether you canceled or returned your order. So the only way it makes any sense is if Amazon is advertising on the chance that you’re going to cancel or return it in the future. Amazon certainly has those numbers. But do you think the chance you’re going to cancel or return it is so high Amazon should advertise the same product to you instead of something else you might buy?
    - grandmczeb 6y ago
      > But do you think the chance you’re going to cancel or return it is so high Amazon should advertise the same product to you instead of something else you might buy? It’s not either or - Amazon showed me grill accessories along with other models when I bought a grill. Regardless, I don’t really have a hard time believing that even a small chance of buying and canceling could make the expected value of showing that item higher than other items, even if that’s counter intuitive to you. I have a harder time believing that what’s shown is not well optimized and “obviously” wrong.
      - danaris 6y ago
        Any individual spot Amazon is advertising you something they could easily know you will not buy is a spot they could have instead used for an ad for something they do not have this knowledge about. So yes, it is either/or.

afiori 6y ago

The best argument for the usefulness/profitability of deep tracking is simple statistical inference. As in if there are 10000 people that with browsing patterns similar to yours and non-trivial fraction of them one day show interest in a discount/service/trip then you are also likely to be interested. It is the official explanation of how facebook/google can magically listen to conversation IRL that they were not supposed to hear: we simply are not as unique/unpredictable as we would like to think.

baybal2 6y ago
I think it's one of the best argument against. It was at least 5 years ago when I started to notice businesses "A/B testing themselves to death" — "It makes no common sense, but increases sales," and then it does until it doesn't, and you are left with no idea what your client actually wants, and no idea how to get out of that "A/B tested to absurdity" situation
- mumblemumble 6y ago
  This also seems to be an inherent danger with thinking that data science == machine learning == AI. I don't want to fundamentally challenge the ŷ-centric approach approach of machine learning. But I do want to say that there's a lot of danger in taking it too far, and thinking that β̂ doesn't matter as long as ŷ is good enough. β̂ has a habit of not mattering until, very suddenly, it really matters quite a lot. And it's even more dangerous to think that an algorithm for generating formulas is somehow doing something intelligent just because you like what you're seeing in the ŷ department.
  - afiori 6y ago
    As someone that is not "in" with those fields what do ŷ and β̂ stand for here?
    - mumblemumble 6y ago
      ŷ is the prediction you get out of the formula, β̂ is the coefficient(s) on the variables in the formula. So, take the simplest possible model: You think that your dependent variable is linear in the dependent variable, and that the line passes through (0,0). That gives us a simplified version of the classic equation for a line, where we say that y is equal to x times some constant that indicates the slope of the line: y = mx In machine learning (or statistical inference), the variable conventions are slightly different, so it would look more like: y = βx And then, since this is machine learning (or statistical inference, or whatever), we don't know the true equation, we're just working with an estimate we've inferred. We stick hats over the estimated values to make it clear that they're estimates: ŷ = β̂x What I was describing, then, is which side of the formula you're focusing on. Traditionally, statistics has been all about scientific inference, and trying to figure out and characterize causal relationships. So you want to know "If I change x in some way, how will that affect the outcome, all else being equal?" β̂ is your tool for teasing those things apart. Machine learning is typically cast as being more about making predictions from observational data. ŷ is just the formal notation for describing those predictions. It's pretty common in machine learning to say, "We are working with observational data, we're not worried about causality, so we'll only pay attention to the predictions." There can be value to that approach, but only to the extent that you really don't care about causality. Which. . . surprisingly often, β̂ matters, even when the people who built the model were trying to ignore it. It becomes especially important when you're using the model to make decisions. For example, when a company is using a machine learning model to try and identify candidates for a position they're trying to fill, they should absolutely be paying attention to what the model thinks about people who liked Lane Bryant on Facebook, and how that compares to people who liked Jos A Bank, and whether they really want things like that to be playing into how they make hiring decisions.
- cortesoft 6y ago
  Yeah, I often wonder this... a/b tests only ever measure short term differences. What if the change you are making increased sales for a few months, but pisses off users in the long run? An A/B test isn't going to show that.
  - shostack 6y ago
    A regression model might if you have sufficient data.
    - cortesoft 6y ago
      You would have to run the a/b test for quite a while, and I don't think many companies do.
code4tee 6y ago
I see that but I think the broader point (and the point of the article) is that too often “success” in finding some less obvious link actually comes at the cost of overlooking the simple stuff. In the end there’s an argument that while neat the fancy ad targeting actually fails to meet the core business objective of maximizing ad performance.
- afiori 6y ago
  My personal opinion is that for most companies having an organically grown audience is more valuable than a lot of faceless traffic. Obviously they can help each other if you do things right and are lucky.

lowwave 6y ago

It totally makes sense that just “dumb” targeting (show people adverts related to the content on the page they are currently looking at, which requires no tracking) might actually work better than all the black magic products companies promote. This is exactly kind of "if all you have is a hammer, everything looks like a nail" that turns me off about AI/ML community.

rhizome 6y ago
Just add 300 DNS queries and wait for the RTB to complete!

Swizec 6y ago

My girlfriend is currently experiencing the dark side of smart tracking. Female who hit 30 and is in a stable relationship. The ML just can’t figure her out. One day every ad is for dating sites, the next day it’s all babies and pregnancy, then engagement rings, wedding locations, more dating sites, a few days of houses to buy ... it’s like the algorithms can’t decide what life stage she’s in.

ceejayoz 6y ago
> it’s like the algorithms can’t decide what life stage she’s in Or, it's working on figuring that out.
- synthc 6y ago
  When my first kid was born, all the ads I got where either baby clothes, daipers etc., or Second Love. A clear separation between A and B
  - mcv 6y ago
    Makes me wonder if there are a lot of people who start cheating once they get a child. What exactly is that algorithm trained on?
- phkahler 6y ago
  Just like she is right? Are you going to marry this guy or move on? Are kids in your future? How about a house, are you two moving in together? Probably all the same crap nosey family and friends want to know. Are advertisers an omnipresent nosey entity? Probably :-)
harry8 6y ago
That's what you want. You don't want these assholes knowing your business. The classic example is when they start advertising a whole lot of pregancy stuff to someone who doesn't want to be pregant and for whom that might be a serious personal problem - advertising seen by others who she does not want to see that. Dumb algos means freedom and privacy. Smart algos that actually work stuff out are dangerous and vile. Nobody gave them permission for that. Nobody complains about seeing ads not specifically tailored for them.

harikb 6y ago

This kind of page-content-based ad targeting did exist way before fine tuned retargeting (say, search or audience/segment retargeting). There are whole bunch of companies that serve this exact market. The reason a more complicated retargeting was created is proof that there is value in it, for advertisers. I am not claiming this is good for people - I am talking only about the business side of things. > don’t need to keep seeing hotel offers for Florida This is not for the lack of smartness in code - it is that the conversion funnel is so narrow that "saving that $30 per month, by not showing you the ad, on a $1m campaign" is not worth tracking and excluding. Although I do believe companies should take the extra effort to avoid the annoyance and be a good citizen - particularly once they have identified "the one real person who actually bought something" among 99 other bots/crawlers/untraceable entities. Not all ad companies appropriately value the real person and the annoyance caused to them. Some do.

mumblemumble 6y ago
Ad companies being able to figure out that I just reserved a hotel room, so they can stop advertising hotels to me, might reduce the annoyance level. But I'm also pretty sure that, if I found out that they had done this, I would never patronize that hotel again.
- harikb 6y ago
  I am talking about cases where the ad results in a click or view through and finally tracked to a purchase from the same provider. You shouldn’t see the same ad again or another ad to rent a hotel room from same hotel chain. I thought that was the case GP was taking about. I am excluding the likes of companies like BlueKai (now Oracle) who facilitate trading the cookie to other companies for further targeting or completely different chain of ads based on a search click to another hotel chain That said, all of this is getting killed by recent Safari and Chrome change of getting rid of all 3rd party cookies
valgaze 6y ago
@harikb, what about something like this: https://i.imgur.com/LYU6NBb.png https://i.imgur.com/LYU6NBb.png Surrounded by content on advice "blue light"-- gets targeted for Blue-tooth lightbulb

Traubenfuchs 6y ago

Amazon showing me a selection of typical buy-once products of the kind I just bought is a depressing sight. They have infinite money and his is the best they can come up with?

grandmczeb 6y ago
I don’t understand why this is such an obviously bad strategy - lots of people buy an item, cancel or return it, and then buy a similar item.
- cgriswald 6y ago
  Let’s say you buy a washing machine on Amazon. You’re not likely to need more than one. Amazon knows whether you canceled or returned your order. So the only way it makes any sense is if Amazon is advertising on the chance that you’re going to cancel or return it in the future. Amazon certainly has those numbers. But do you think the chance you’re going to cancel or return it is so high Amazon should advertise the same product to you instead of something else you might buy?
  - grandmczeb 6y ago
    > But do you think the chance you’re going to cancel or return it is so high Amazon should advertise the same product to you instead of something else you might buy? It’s not either or - Amazon showed me grill accessories along with other models when I bought a grill. Regardless, I don’t really have a hard time believing that even a small chance of buying and canceling could make the expected value of showing that item higher than other items, even if that’s counter intuitive to you. I have a harder time believing that what’s shown is not well optimized and “obviously” wrong.
    - danaris 6y ago
      Any individual spot Amazon is advertising you something they could easily know you will not buy is a spot they could have instead used for an ad for something they do not have this knowledge about. So yes, it is either/or.