4 ms·

A massive misallocation of online advertising dollars

44 points by prakash 17y ago

vinhboy 17y ago
This is why Amazon is great. When people from the internetz click through to Amazon, they get all the reviews/research they need. They are also presented with other great items they didn't even intend to buy. So that means profit for content creators. But I think the biggest winners are deal sites like slickdeals.
- patio11 17y ago
  But I think the biggest winners are deal sites like slickdeals. The biggest winner by FAR in "last click wins" is Google. Brand queries outconvert just about everything else, because they frequently happen late in the sales cycle after the customer has picked out what they want to buy and now just need navigation to the merchant. All those little yellow boxes -- the ones many Google searchers think mark the extra quality results (no joke!) -- mean Google gets paid money to parasite off the brand query. Unsophisticated stats monkeys might look at that and say "Dang, Google CPC ads for the win again!"
  - ntoshev 17y ago
    All those little yellow boxes -- the ones many Google searchers think mark the extra quality results (no joke!) Citation needed.
IgorPartola 17y ago
This is very interesting. I have always viewed content sites (especially blogs) as a sort of parasite. Here's the manufacturer/seller that is trying to sell me something that I want and in the way there are numerous blogs that take away the seller's money onto advertising, therefore raising my cost. This of course only applies to the sites I don't read. There are legitimate ones, ones that do the research, publish their results, etc. Edmunds.com falls under this category. On the other hand, there is a metric ton of travel sites and it's just a huge waste of my time. Every time I want to travel I have to check 5-6 sites to see if I'm getting a good deal or not. Why can't I go to the source and get data directly from the airlines?
johnrob 17y ago
If brand advertising on content sites is under valued, isn't that a big opportunity for brands to buy valuable ad space on the cheap? The author doesn't explain why the market is wrong (I'm of course skeptical, because markets aren't wrong very often).
- lmkg 17y ago
  The issue is availability of information. The further from the point of purchase you get, the more difficult it is to properly credit parts of the funnel with driving purchase intent. Most analytics tools don't do a good job of crediting multiple parts of the funnel for the purchase attempt. Google Analytics, for example, can only credit one campaign per conversion event, which by default is the most recent one (it can be configured to the first one, within a time window). Engagements earlier in the funnel that drive purchase intent aren't properly credited, so they look like a less efficient place to allocate funds than the really are. The result is that resources are being allocated correctly based on the information, but the information is wrong, so as a whole the market is inefficient. Ad allocation tends to be relentlessly data-driven in optimizing, but also rather heads-down: people trust the data too much, without necessarily recognizing the flaws of the tools.
aresant 17y ago
My counter argument: Since the beggining of time consumers have undergone the same process when making a buying decision. In the past you might drive to the convenient local Circuit City to check out a TV you read in consumer reports was quality, and then order the thing from Sears catalog because you were discount shopping. I agree that it's a shame we can't all split a few pennies of advertising dollars on the research side, but I look at the REAL opportunity here as helping people create BETTER websites & CRM to shorten the sales cycle and capture those dollars for themselves.
- minouye 17y ago
  I'd agree with this and add that merchants/affiliate networks should further encourage better publisher sites by measuring and paying out on metrics around the lifetime value of a customer. A better customer experience on an affiliate site will drive repeat purchases and ultimately the lifetime value of the customer will be greater than the margins on the initial sale. It would be great if quality publishers could be compensated for "better" customers, because it would reward valuable affilates and cut down on a lot of the thin landing pages that currently plague Google on long-tail keyword searches.
  - qeorge 17y ago
    If you have high-quality traffic which generates a lot of sales, you actually can negotiate better rates with affiliate networks. Their published payouts are just a starting point for the negotiation.
johnrob 17y ago
Another counter argument: When you target earlier in the funnel, you're less sure of winning that user. Example: you buy a banner ad promoting a digital camera. The user sees it, and decides it's time to buy one. Now he does research, and ends up buying a competitor. While you were successful in creating demand, you ended up sending a customer to a competitor. That is why display ads are worth less - you aren't guaranteed to extract the value they create.
- ntoshev 17y ago
  I thought the same. However some content would be better for you and some content would be better for your competitors, and the advertising budgets should be allocated according to the generated sales. Suppose the content in question is an article on high speed photography, and some cameras are better at this than others. Surely an efficient market would reward the article with proportionally more advertising dollars from high speed camera sellers.

44 points by prakash 17y ago

This is why Amazon is great. When people from the internetz click through to Amazon, they get all the reviews/research they need. They are also presented with other great items they didn't even intend to buy. So that means profit for content creators. But I think the biggest winners are deal sites like slickdeals.

patio11 17y ago
But I think the biggest winners are deal sites like slickdeals. The biggest winner by FAR in "last click wins" is Google. Brand queries outconvert just about everything else, because they frequently happen late in the sales cycle after the customer has picked out what they want to buy and now just need navigation to the merchant. All those little yellow boxes -- the ones many Google searchers think mark the extra quality results (no joke!) -- mean Google gets paid money to parasite off the brand query. Unsophisticated stats monkeys might look at that and say "Dang, Google CPC ads for the win again!"
- ntoshev 17y ago
  All those little yellow boxes -- the ones many Google searchers think mark the extra quality results (no joke!) Citation needed.

IgorPartola 17y ago

This is very interesting. I have always viewed content sites (especially blogs) as a sort of parasite. Here's the manufacturer/seller that is trying to sell me something that I want and in the way there are numerous blogs that take away the seller's money onto advertising, therefore raising my cost. This of course only applies to the sites I don't read. There are legitimate ones, ones that do the research, publish their results, etc. Edmunds.com falls under this category. On the other hand, there is a metric ton of travel sites and it's just a huge waste of my time. Every time I want to travel I have to check 5-6 sites to see if I'm getting a good deal or not. Why can't I go to the source and get data directly from the airlines?

johnrob 17y ago

If brand advertising on content sites is under valued, isn't that a big opportunity for brands to buy valuable ad space on the cheap? The author doesn't explain why the market is wrong (I'm of course skeptical, because markets aren't wrong very often).

lmkg 17y ago
The issue is availability of information. The further from the point of purchase you get, the more difficult it is to properly credit parts of the funnel with driving purchase intent. Most analytics tools don't do a good job of crediting multiple parts of the funnel for the purchase attempt. Google Analytics, for example, can only credit one campaign per conversion event, which by default is the most recent one (it can be configured to the first one, within a time window). Engagements earlier in the funnel that drive purchase intent aren't properly credited, so they look like a less efficient place to allocate funds than the really are. The result is that resources are being allocated correctly based on the information, but the information is wrong, so as a whole the market is inefficient. Ad allocation tends to be relentlessly data-driven in optimizing, but also rather heads-down: people trust the data too much, without necessarily recognizing the flaws of the tools.

aresant 17y ago

My counter argument: Since the beggining of time consumers have undergone the same process when making a buying decision. In the past you might drive to the convenient local Circuit City to check out a TV you read in consumer reports was quality, and then order the thing from Sears catalog because you were discount shopping. I agree that it's a shame we can't all split a few pennies of advertising dollars on the research side, but I look at the REAL opportunity here as helping people create BETTER websites & CRM to shorten the sales cycle and capture those dollars for themselves.

minouye 17y ago
I'd agree with this and add that merchants/affiliate networks should further encourage better publisher sites by measuring and paying out on metrics around the lifetime value of a customer. A better customer experience on an affiliate site will drive repeat purchases and ultimately the lifetime value of the customer will be greater than the margins on the initial sale. It would be great if quality publishers could be compensated for "better" customers, because it would reward valuable affilates and cut down on a lot of the thin landing pages that currently plague Google on long-tail keyword searches.
- qeorge 17y ago
  If you have high-quality traffic which generates a lot of sales, you actually can negotiate better rates with affiliate networks. Their published payouts are just a starting point for the negotiation.

johnrob 17y ago

Another counter argument: When you target earlier in the funnel, you're less sure of winning that user. Example: you buy a banner ad promoting a digital camera. The user sees it, and decides it's time to buy one. Now he does research, and ends up buying a competitor. While you were successful in creating demand, you ended up sending a customer to a competitor. That is why display ads are worth less - you aren't guaranteed to extract the value they create.

ntoshev 17y ago
I thought the same. However some content would be better for you and some content would be better for your competitors, and the advertising budgets should be allocated according to the generated sales. Suppose the content in question is an article on high speed photography, and some cameras are better at this than others. Surely an efficient market would reward the article with proportionally more advertising dollars from high speed camera sellers.