Are banners/sponsored posts more efficient than purchased keyword? I would have thought someone searching on google wants to click, that's why he is on google for. Someone checking what their friends are up to on facebook is merely annoyed by unrelated content.

I don't see FB/G as direct competitors.

Google is good to sell products that are actively searched for.

People go to Facebook when they are bored. Facebook is great to sell products to people who are bored (various forms of entertainment products and services). It's also good to sell stuff people aren't actively searching.

So, depending on the product/service you are offering you can pick a different channel to reach your customers. There's no one-size-fits-all solution.

Sometimes I click on those purchased keywords by mistake. Then get presented with the inevitable "could not connect to server" message because they use tracking servers which are blocked at the router level.

Main problem with advertising in general is more and more people are learning it's best use is as a signal of what not to buy.

Facebook on the other hand. At least can pretend it's something a friend is recommending. Although that is transparent enough to be even worse.

Google has context. You search for something, chances are you're interested in buying it. Placing an ad in the search page has better ROI for advertisers. At Facebook you can build loyalty which is a different beast altogether.

So is it safe to say that Google is better for acquiring hot leads, while Facebook is better for branding?

Yes - also important to remember that Facebook has Instagram too, which is a nicer (imo) ad format and self selects for more affluent users.

That sounds right.

Along those lines, Amazon is a much more dangerous competitor to AdSense than Facebook is at the moment.

This is reliant on people continuing to use Facebook as their primary social networking tool.

In my environment (Netherlands) I'd say that WhatsApp is now the primary social networking tool for most people. I wonder how that is going to work out for Facebook.

Facebook owns WhatsApp. The question is what they are going to do with it. WhatsApp with end to end encryption seems to not allow for the type of monetization people are talking about here.

That's what I mean: how will they monetize it? They'll probably introduce business / bot accounts that they can monetize, but the encryption makes all sorts of monetization schemes hard - which is what I like about it.

Maybe it's possible to do ad-targeting on the client.

Facebook bought Whatsapp two years ago.

What's done is done. Facebook won't delete your data, once U registered, give your gender/age, you are probably forever screwed, forever becoming a data point that waiting to be targeted.