I think the answer is that we believe SN sites should be able to make more money than these other businesses. The person who figures out how to make that happen will be rich, but I don't think the answer is to just add another business onto your SN.
I should have made more clear that this is a longer term strategy, and yes, indeed more expertise is needed in these industries. However, what these industries do is not rocket science, and can be reproduced over time.
My point is that SN sites could eventually do the conversion of that sort of traffic themselves, and gain higher CPM In the Long Run. This process would take years, but is rewarding in the end.
He has two assumptions:
- You can make a profit in the industry of your advertisers. This is a huge if, and is impacted by scale, expertise, management focus, time/maturity, branding and many other factors.
- The return on investment you can make in industry X is higher than the return on investment you can make by expanding your operations in your current industry. This could be true for social sites, where increased investment (i.e. programmer time) might not get them many more users, but they still have to compare the possible revenue to the increased profit from firing some programmers and going into maintenance mode.
These to are the main reasons why you see companies outsourcing non-core activities (like cleaning and even product development), the markup they now have to pay on the service is lower than their internal cost of capital.
-That return on investment would be higher, in the long run. Yes, thats an if, but its a much better if than "social graphs completely turning around CPM for something like FB".
Those are also, by their opposite, reasons why companies consolidate and acquire...
For an example of conflicts of interest look at the Windows/Office vertical integration and the constraints placed on Office in order to keep Windows profits high.
Your second point boils down to the fact that social networks give an awful return on investment. Could be true, but if so that's more of an argument to get out of social networks altogether and find a more profitable business.
Also, if mortgage brokers are only willing to pay some portion of 14 cents CPM there, that means Facebook is creating very little value for them. Which also means that if Facebook got into the mortgage selling business, they could probably capitalize off of it in much better ways than just advertising it on their relatively worthless site.
So in essence, the only industries Facebook would benefit from entering are ones that already pay them a high CPM. But if people were paying them a high CPM, they wouldn't be in this predicament.
The site is indeed relatively worthless, and I doubt anybody at all is paying them a high cpm in anything. Thats the whole point, to turn low cpm into somewhat higher cpm, in the long term by capturing more of the revenue stream.
Is it fool proof? No way. Facebook or whatever, could risk losing their investment in whatever company they bring up, and also lose the opportunity cost on whatever ads they could be selling instead of sending to their own child company instead... But does it stand a better chance at making more money in the long run compared to magical BS like social graphs and social ads? Yes, I definately think so.
I don't think you're understanding why this provides no value to anyone. You're turning Facebook into a Berkshire Hathaway of sorts, with the only coherency being that every company they buy a stake in advertises on Facebook. By that logic, why shouldn't Time Warner start buying stakes in whoever advertises on CNN? Why shouldn't everyone who sells ads do this?
Facebook gains no higher CPM from doing this either. I mean, they may make more money if these auxiliary businesses do well. But it's no more than if they kept selling ads on Facebook for 14 cents cpm, set up these side companies, and spent 14 cents cpm advertising them somewhere else. The fact that they own Facebook provides no value whatsoever for the side companies.
Effectively you're saying that any company with a non-functioning business model should just do something else too. That doesn't solve the problem.
