Comments on: Real-time bidding sounds painful

By: The Real Costs of Real-Time Bidding (RTB)

The Real Costs of Real-Time Bidding (RTB) — Thu, 10 Dec 2009 06:23:19 +0000

[...] Right Media product manager (and a former colleague of mine and disclosure an advisor to CPMa) blogged it in September 2009 which I will reproduce here verbatim as a nice summary of some of the issues: This sounds pretty [...]

By: Mike On Ads » Blog Archive » RTB Part I Followup

Mike On Ads » Blog Archive » RTB Part I Followup — Sat, 19 Sep 2009 19:14:46 +0000

[...] so in the last post I explained what Real Time Bidding (RTB) was. Greg Yardley wrote a response on his blog which made me realize that I didn’t spend quite enough time on the technical implications of [...]

By: Brent

Brent — Sun, 06 Sep 2009 22:09:38 +0000

RTB is huge for people because it disintermediates big networks, if one were so inclined. The cost of decisioning hundreds of impressions to find the one you want is a lot less than the cost of buying hundreds of impressions to find the one you want. The result is that, to cherry-pick a lot of inventory today, you have to be a big network. That is why, when people looked at the ad network market prior to RTB, it was a natural monopoly. The guy with the most advertisers paid the most for impressions, the guy who paid the most got the most inventory, the guy with the most inventory gets the most advertisers, etc..

With the advent of RTB, people can pay to cherry-pick impressions in front of the network. That is enormous!

By: Jerry

Jerry — Sat, 05 Sep 2009 16:49:37 +0000

In the spirit of creating products to serve your customers, I’ve been scratching my head over which buy-side customers really want RTB (aside from the lead generators and other bargain hunters.) Can you imagine any of the big media buying shops having expertise bidding in real-time? Maybe the VMMs and CPMa’s of the world, but the guys who learned media three-martini-lunching the upfront?

On the publisher side, not nearly enough liquidity for there to be an efficient market in real-time. Usually, no liquidity means a huge bid-ask spread, but with perishable inventory low-bids clear. Publishers who sign up for this are going to compound the problems they’ve already sold themselves into: selling inventory to bargain hunters at bargain prices (and then getting snubbed by media buyers who know they can find inventory at cut-rates.)

The idea that our five year old, few billion dollar online media marketplace needs the same type of infrastructure as our centuries old, multi-trillion dollar financial markets seems a bit fantastic. What they really need is tools that better maximize the ‘R’ of the advertiser’s ROI (rather than the ‘I’) and the ‘C’ of the media’s CPM (rather than the ‘M’.)

All IMHO, natch.

By: John Ebbert

John Ebbert — Sat, 05 Sep 2009 15:56:33 +0000

I agree but…. How about this – I think conversion reporting does not necessarily need to be real-time to take advantage of the benefit of real-time, impression-level bidding.

If, in real-time, you can bring your own data and analyze what the impression is worth to you using predictive models on conversion metrics – well then, this is a lot better than what’s going on today where plenty of waste occurs when buckets of impressions are bought in hopes of finding a couple that might work.

By: Brian O'Kelley

Brian O'Kelley — Sat, 05 Sep 2009 12:55:08 +0000

Greg – great points, and you’re right on all of them. I would expand the conversion lag challenge to include attribution in general. If your attribution model doesn’t work, you’re screwed in either a traditional or RTB world.