3 ms·

This is why I believe ad tech is a zero-sum game. New channels (banner ads, social ads, online video ads, etc) appear ad old ones dry (TV, radio, etc) up. But

This is why I believe ad tech is a zero-sum game. New channels (banner ads, social ads, online video ads, etc) appear ad old ones dry (TV, radio, etc) up. But the spend doesn't -- an ad buyer has a campaign budget of $X, and is looking to maximize the amount of lift they can get for that budget.

Denzel 9y ago
You are correct. See: https://www.bloomberg.com/news/articles/2014-03-03/advertisings-century-of-flat-line-growth https://www.bloomberg.com/news/articles/2014-03-03/advertisi.... Ad spend as a % of U.S. GDP remains relatively constant since the 1920s.
carapace 9y ago
"50% of your advertising budget is wasted, the problem is no one can tell you which 50%." I don't think that's changed?
- exelius 9y ago
  It's changed a little bit -- 50% of your advertising budget is wasted, and you can generally tell which 50% -- but past performance is now a poor indicator of future returns. So you can tell what 50% was wasted, but it's harder to tell what's going to work next time.
  - shostack 9y ago
    The question of which 50% gets even murkier the more data we have. For example, cross-channel/cross-device attribution used to be a known thing, but it was always pretty hazy because pretty much all analytics and conversion tracking was last click (or first touch if you're a direct mail company). Then we started getting attribution tools and getting different pieces of the puzzle. Now the issue has become that there's all these ways of looking at the data that can point you in the right general direction, but short of dynamic/data-driven attribution models with MASSIVE amounts of data feeding into them, it can still be difficult to say "actually this is what this channel did for us." For orgs comfortable with data that is a bit fuzzier, that's fine, but for orgs where people are held to strict numbers based on the last click, it can hobble them when it comes to higher-funnel or view-through heavy channels like social, display, PR, etc.
shostack 9y ago
As an ad buyer who has done so both at agencies and client-side, I'll say that this is true to an extent, but that maximizing lift isn't always the priority. For example, I may be able to eek out a small incremental gain in performance on some tiny network or via a deeply-integrated direct buy, but how much time would that suck up? What would I need to do from a tracking perspective? How much would it cost to get all of the necessary creative? What about reporting? Does the data get pulled into my existing reporting automatically, or will I need to perpetually marry up some extra data set in Excel and dramatically increase reporting time? What about time for ongoing management that happens outside of my existing tools? These are all important considerations. Actual channel performance is one of them, but far from the only or even the most important depending on circumstances.

Denzel 9y ago

You are correct. See: https://www.bloomberg.com/news/articles/2014-03-03/advertisings-century-of-flat-line-growth https://www.bloomberg.com/news/articles/2014-03-03/advertisi.... Ad spend as a % of U.S. GDP remains relatively constant since the 1920s.

carapace 9y ago

"50% of your advertising budget is wasted, the problem is no one can tell you which 50%." I don't think that's changed?

exelius 9y ago
It's changed a little bit -- 50% of your advertising budget is wasted, and you can generally tell which 50% -- but past performance is now a poor indicator of future returns. So you can tell what 50% was wasted, but it's harder to tell what's going to work next time.
- shostack 9y ago
  The question of which 50% gets even murkier the more data we have. For example, cross-channel/cross-device attribution used to be a known thing, but it was always pretty hazy because pretty much all analytics and conversion tracking was last click (or first touch if you're a direct mail company). Then we started getting attribution tools and getting different pieces of the puzzle. Now the issue has become that there's all these ways of looking at the data that can point you in the right general direction, but short of dynamic/data-driven attribution models with MASSIVE amounts of data feeding into them, it can still be difficult to say "actually this is what this channel did for us." For orgs comfortable with data that is a bit fuzzier, that's fine, but for orgs where people are held to strict numbers based on the last click, it can hobble them when it comes to higher-funnel or view-through heavy channels like social, display, PR, etc.

shostack 9y ago

As an ad buyer who has done so both at agencies and client-side, I'll say that this is true to an extent, but that maximizing lift isn't always the priority. For example, I may be able to eek out a small incremental gain in performance on some tiny network or via a deeply-integrated direct buy, but how much time would that suck up? What would I need to do from a tracking perspective? How much would it cost to get all of the necessary creative? What about reporting? Does the data get pulled into my existing reporting automatically, or will I need to perpetually marry up some extra data set in Excel and dramatically increase reporting time? What about time for ongoing management that happens outside of my existing tools? These are all important considerations. Actual channel performance is one of them, but far from the only or even the most important depending on circumstances.