It is clear that depending on the shape of the demand curve, decreasing supply may even increase revenue. So obviously, publishers are motivated to control the supply.

this fits in nicely with the fact that there is a natural segmentation in the market between direct response and premium – and they have different demand curves. so the aggregate supply pool can be broken into two pools such that the overall revenue is maximized.

i think i am stating the obvious (and i see u are pointing to the Adotas articles as well – which are saying the same thing).

but good series of posts nevertheless!

The ‘premium’ stuff is generally reserved in advance and is as close to guaranteed as possible – the contracts for buying ‘premium’ all have punitive ‘make-good’ clauses if the publisher doesn’t deliver. The ‘remnant’ stuff might be purchased in advance, or it might be auctioned off in real-time, but the publisher doesn’t face the threat of make-goods if he doesn’t deliver.

As you might expect, this advance reservation comes at a price – the advertiser pays a ‘premium’ for it. But my theory is that this is fine for the big agencies, who are motivated to spend *all* of their advertising budgets (if you don’t spend it you’ll lose it next time around) and who don’t want to deal with the risk that they might spend less than their allocated budgets.

So, definitely ‘premium’ vs. ‘remnant’ is misnamed, and it probably should be renamed something like ‘reserved’ vs. ‘pre-emptible’. (Or, if you like Wall Street analogies, ‘forward contract’ vs. ‘spot market’.) But I still think there’s a useful distinction somewhere in there.