“But display ads lift your other efforts”
Display ads don’t usually work in isolation (i.e., on a click to purchase model).
However, several studies have shown that display ads result in more searches on branded terms and more visits to your website. The display ad salesman will tell you that if you’re making a profit in those areas — and if you’re not you might be in the wrong business — then display ads could make up for their dismal click-to-purchase perfomance by helping your channels that are making money.
The problem with this kind of analysis is that it doesn’t account for how much you have to spend on display ads to get the lift.
For example, just to make the numbers easy, let’s say you get 100 “direct” visits to your site every day (from people who type in your address), 100 visits from natural search results and 100 visits from paid search results.
Let’s say you can convert 5% of your visitors into $10 sales. That gets you $150 a day in revenue. (5% of 300 visits times $10.)
Of course you’ll have different numbers for the different sources, but let’s keep this easy.
Now let’s say that a display ad campaign can increase each of these by 50%. So now you’re getting $225 in revenue. (5% of 450 visits times $10.)
If you spent less than $75 on your display ad campaign, you’re doing well. If you spent more, maybe not. You have to look at the numbers more carefully than we’re doing here.
The point is that the simple claim that display ads increase search by 50% doesn’t tell me anything useful. How much do I have to spend on display ads to get that 50% lift?
What we really need is a study that compares PPC spend with display ad spend. IOW, what percent of your PPC budget do you have to spend on display ads to get what percent lift?