3 ms·

There's a lot of interesing stuff in this comment. But it seems to conflate lead-gen with advertising. There are lots of reasons to advertise that don't relate

by 001sky 12y ago

There's a lot of interesing stuff in this comment. But it seems to conflate lead-gen with advertising. There are lots of reasons to advertise that don't relate to the sales funnel. The second issue is alot of internet advertising is bullshit, including the metrics. Now, some things are good--actual-purchase links are decent data, but again that's really lead-gen not advertising. If your lead already wants to buy your product, its too late for advertsing and you're really more into a different game. Nobody looks at an ad in the superbowl and the runs out and buy something 12 minutes later. The fact your narrative doesn't apply to that case is more a sign that something is missing in it, not that superbowl ads are mis-priced per se.

sbenario 12y ago
The metrics are often broken. Think about how you measure it... Ideally, you want to pay the ad platform for sending you a user who will buy your product. So you say something like: "I'll pay you X for every user you send me who buys something". Great, right? Except now the ad can simply optimize for users who were GOING to buy your product anyway. So now you've spent money getting users who would have been there anyway. What you really want is to pay for users "Who buy my product, but would NOT have bought it without the presence of the ad". And that's a much harder thing to measure.
- ghshephard 12y ago
  "Ideally, you want to pay the ad platform for sending you a user who will buy your product." Brand advertising is built around the concept of "When a person is making a buying decision, I want them to buy my product." We know it works, because when you stop doing brand advertising for things like Toilet Paper, Detergent, Toothpaste, Deoderant, etc... sales go down. And when you start the advertising again, sales go up.
  - mjevans 12y ago
    Don't people just sample the market to find out a brand that is to their tastes (performs as desired) and then keep buying it while the price is still of value?
    - ghshephard 12y ago
      There are a lot of theories, but, the one thing that has been consistent for at least the last 60 years, is that if you have a consumer Brand of something, say "Tide Detergent" - and you stop all your brand advertising in a certain region, sales go down in that region, and when you start advertising again, sales go back up. Price, Value, Placement, and, obviously, competition are all conflating factors as well - but the underlying response to advertising has been pretty consistent.
    - mkr-hn 12y ago
      I've always wondered what the marketing campaigns look like for the 100 different brandings of the same product at the grocery store. Do Toothpaste A and Toothpaste Z340 target completely different demographics, or are they going for the same eyeballs with their ads?

by 001sky 12y ago

sbenario 12y ago

The metrics are often broken. Think about how you measure it... Ideally, you want to pay the ad platform for sending you a user who will buy your product. So you say something like: "I'll pay you X for every user you send me who buys something". Great, right? Except now the ad can simply optimize for users who were GOING to buy your product anyway. So now you've spent money getting users who would have been there anyway. What you really want is to pay for users "Who buy my product, but would NOT have bought it without the presence of the ad". And that's a much harder thing to measure.

ghshephard 12y ago
"Ideally, you want to pay the ad platform for sending you a user who will buy your product." Brand advertising is built around the concept of "When a person is making a buying decision, I want them to buy my product." We know it works, because when you stop doing brand advertising for things like Toilet Paper, Detergent, Toothpaste, Deoderant, etc... sales go down. And when you start the advertising again, sales go up.
- mjevans 12y ago
  Don't people just sample the market to find out a brand that is to their tastes (performs as desired) and then keep buying it while the price is still of value?
  - ghshephard 12y ago
    There are a lot of theories, but, the one thing that has been consistent for at least the last 60 years, is that if you have a consumer Brand of something, say "Tide Detergent" - and you stop all your brand advertising in a certain region, sales go down in that region, and when you start advertising again, sales go back up. Price, Value, Placement, and, obviously, competition are all conflating factors as well - but the underlying response to advertising has been pretty consistent.
  - mkr-hn 12y ago
    I've always wondered what the marketing campaigns look like for the 100 different brandings of the same product at the grocery store. Do Toothpaste A and Toothpaste Z340 target completely different demographics, or are they going for the same eyeballs with their ads?