TV Measurement Is A Chaotic Mess; First Price Auctions Comes With Risks

Here’s today’s AdExchanger.com news round-up… Want it by email? Sign up here.

Gold Mine Of Information

People are watching more television, but less measurably. “We know the audience is there,” GroupM US managing director Ed Gaffney tells Mike Shields for a Business Insider on the chaotic state of TV measurement. “TV is not dying, it’s evolving. And we really want to evolve with it. … The problem is, everybody has their own data and nobody shares.” For its Olympics advertisers, NBC is building on the standard Nielsen fare with data from Adobe, FreeWheel and Oracle-owned Moat. And other networks will be paying close attention. “We are still measuring at the speed of the horse and buggy,” says Discovery Networks CEO Jean-Briac Perrette. “The Olympics are probably the best laboratory for multi-screen usage. So we cannot just sit by. We’re not going to keep waiting for third parties to get this together.” More.

At First Glance

Publishers are the short-term winners from the shift to first-price auctions, since they’re paid the highest bid price. “Over time, however, the forces of auction dynamics come into play,” Kerry Bianchi, CEO of Visto (née Collective), writes at MediaPost. Bidding on different models across open exchanges could confuse or frustrate advertisers, at a time when it’s all too easy to consolidate spend with walled gardens. Buyers may drastically reduce bid prices to “test price elasticity.” More.

Garbage Out

Lotame purged 10% of its third-party data exchange after discovering 400 million profiles were bots and fraudulent accounts. Lotame, one of the largest data exchanges on the market, has been helped by bot fraud detection startup Distil Networks. GDPR and the increased ability to attribute real-world campaign value are pressuring ad tech companies to jettison the junk. “First and foremost, advertisers need to trust the quality of the data and then look at the scale,” said Michelle Mirshak, VP of data architecture and platforms at Spark Foundry. “Scale with a disregard for quality and transparency is not acceptable.” More at Adweek.

Incoming

Facebook patented a method to segment users based on socioeconomic class. It’s an algorithm that factors in home ownership status, education, mobile and desktop device usage and data from the social platform to bracket people into three income tiers. Read the patent. Facebook came under heat last year for enabling illegal race-based targeting practices for housing and employment notices. And while socioeconomic segmentation isn’t technically illegal, Facebook’s inexact way of bucketing users could raise eyebrows. More at Engadget.

But Wait, There’s More!