jump to navigation

Data exhaust moves beyond targeted marketing and into financial services decision making November 27, 2010

Posted by jeremyliew in data, financial services, marketing, targeting.
2 comments

Fascinating article in the WSJ a couple of weeks ago about how the big insurance companies are testing using data profiles to identify risky clients. Using the data is potentially an alternative to the costly physical exams currently used to underwrite health insurance policies:

 

In one of the biggest tests, the U.S. arm of British insurer Aviva PLC looked at 60,000 recent insurance applicants. It found that a new, “predictive modeling” system, based partly on consumer-marketing data, was “persuasive” in its ability to mimic traditional techniques…

Making the approach feasible is a trove of new information being assembled by giant data-collection firms. These companies sort details of online and offline purchases to help categorize people as runners or hikers, dieters or couch potatoes.

They scoop up public records such as hunting permits, boat registrations and property transfers. They run surveys designed to coax people to describe their lifestyles and health conditions.

Increasingly, some gather online information, including from social-networking sites. AcxiomCorp., one of the biggest data firms, says it acquires a limited amount of “public” information from social-networking sites, helping “our clients to identify active social-media users, their favorite networks, how socially active they are versus the norm, and on what kind of fan pages they participate.”…

Acxiom says it buys data from online publishers about what kinds of articles a subscriber reads—financial or sports, for example—and can find out if somebody’s a gourmet-food lover from their online purchases. Online marketers often tap data sources like these to target ads at Web users.

Not everyone is comfortable with this approach. Some regulators have raised potential concerns:

“An insurer could contend that a subscription to ‘Hang Gliding Monthly’ is predictive of highly dangerous behavior, but I’m not buying that theory: The consumer may be getting the magazine for the pictures,” says Thomas Considine, New Jersey’s commissioner of banking and insurance.

I think I’d bet against Mr. Considine on this one.

I’m fascinated by the idea of using publicly available data to make better underwriting decisions, whether for insurance or for lending. This isn’t a new idea. Student loans first became a growth industry when someone decided that using a students major (pre-med vs liberal arts) or GPA could help them decide who to lend to and how much. But as the amount of data available has exploded, whether directly reported (e.g. on social networks), inferred from behavior (e.g. web surfing and ecommerce habits) or volunteered as part of an application (e.g. bank account log in info, as supplied to Mint, that can show regularity of income and cash payments), a financial instituions ability to underwrite more individually goes well beyond FICA scores.

I’m interested in any companies looking at doing something like this. Email me.

 

Online direct response shows booms and busts by category January 30, 2009

Posted by jeremyliew in advertising, targeting.
add a comment

Earlier this month the Lookery blog looked at the booms and busts of behavioral targeting.

They point out that behavioral targeting is most visibly effective for online direct response in considered purchases (where the decision to buy takes a while, so there is some time to notice purchase intent and start to target advertising). They also point out that some of the industries that have these characteristics have gone through wild cycles of increased then decreased demand, sometimes for structural reasons, sometimes for legal reasons, and sometimes for economic cycle reasons:

online direct response booms and busts

online direct response booms and busts

* Auto – 2004-2008
Auto has always been a major driver of behavioral targeting the last few years benefitting everyone from Google to Internet Brands (aka Carsdirect.com)
* Household Durables – 2004-2008
The rise of high end e-commerce fueled consumer electronics and appliance e-commerce sties
* Mortgage & Finance – 2003-2006
With qualified leads selling from $50-500 a pop, companies like LowerMyBills made a killing
* Mobile Phones & Ring Tones – 2004-2006
A key driver of under 30 monetization as mobile co’s and ring tone services went wild chasing this high converting group
* Online Dating – 2004-2006
Before the rise of free sites like OkCupid, Plentyoffish, Woome, & Craigslist, dating paid some of the best CPA’s online
* Online Gambling 2002 to 2005
Gambling esp. poker was a major gold rush for online pubs emerging from the last ad recession with high payouts at least until the FTC decided to get all uppity

The economic cycle will turn eventually and bring back auto, mortgage and durables demand. Mobile and Online Dating may not return as strongly as they have been pressured by prices falling to marginal cost. I think online gaming will drive the next cycle. Let’s see.

Don’t count on ad targeting to lift CPMs in the near term October 2, 2008

Posted by jeremyliew in advertising, contextual targeting, targeting.
3 comments

The online ad market is not immune to the overall advertising recession, and growth has slowed. Many online media companies and ad networks are counting on targeting to help lift CPMs.

But Julie Ruvolo reports from the Adweek conference last week that media buyers are still hesitant about highly targeted ad campaigns:

In the traditional media-buying paradigm, advertisers buy audiences by buying content. Coca-Cola sponsors American Idol, Nissan sponsors Heroes, and so forth. But social media, ad networks, and especially behavioral ad networks, are chipping away at the “content as a proxy” mentality and positing that ads can be as or more effective if they’re tied directly to people and not to content…

But for all the talk it’s garnering, media buyers remain hesitant about jumping on the addressability band-wagon for several reasons.

First, while agencies are opening up to a more data-centric approach, operational challenges abound. One of the key issues is that it’s easier to buy a national TV ad than to set up and constantly manage a million-word AdSense campaign, or develop video creative for hundreds of demographics instead of one broad demographic…

Further, advertisers are struggling with the sheer volume and sophistication of data available to them. As digital marketing agency Avenue A’s Andy Fisher said, “We’re drowning in data.” …

We can talk all day about how wonderful digital media is, how addressable and trackable and cheap the media is, but the reality is that there’s a decades-long and multi-billion-dollar symbiosis between the ad industry and the TV industry. It’s going to take more than superior product, logic and efficiency to supplant that relationship.

I think Ruvolo is right.

Online advertising has typically been sold in one of three ways:

1. Endemic advertising targeted against relevant content, typically commanding double digit CPMs. An example would be selling movie advertising against Flixster (a Lightspeed portfolio company).
2. Demographically targeted advertising, typically targeted against relevant content, typically commanding low single digit CPMs. An example would be selling a “women” demo against TMZ.
3. Broad reach inventory, typically commanding $1 or below CPMs. An example would be a selling Yahoo email inventory.

Advertisers are comfortable with buying advertising against content adjacencies.

There are four flavors of ad targeting popular today:

1. Geographic
2. Demographic
3. Contextual
4. Behavioral

Through demographic and behavioral targeting, online media companies are asking advertisers to follow the user instead of following the content:

But, online ads should follow users and communities, since users are the ones to decide what content they want to put where, says David Carlick, Managing Director at Vantage Point Venture Partners.

“[I] say no, now you [the advertiser] are sponsoring the consumer—not the content online, but what they want to do online. If they want to go on MySpace and look at half-naked drunk photos, who are you to say that’s not good for my brand? You need to go where the people are and sponsor what they do, and not attach yourself to the 5% of content that looks like TV.”

Or as Jeff Jarvis says:

That’s [buying content adjacencies] still treating us like a mass. That’s still about lazy advertisers who want to buy upfront and don’t want to converse with us as individuals or at least communities. We need advertisers’ money; that will be the primary support of online media. But we need to both retrain them and give them the infrastructure and data to enable them to market smarter and create meaningful relationships — and, in the process, support small instead of big.

In my experience, when the guys with the money [advertisers] want to do things one way, and the guys who want the money [media companies] want to do things another way, then it is usually the guys with the money who walk away happy.

Behavioral and demographic targeting to the user level will likely have success with direct response advertisers who can readily measure and potential lift in response rates. But brand advertisers will want to continue doing business the way they are used to doing business. Furthermore, an advertising recession is not going to be an easy time to “retrain” advertisers. Content adjacencies will likely be the way most brand advertising is sold for the next couple of years at least.

Four flavors of ad targeting July 7, 2008

Posted by jeremyliew in ad networks, advertising, targeting.
9 comments

I recently had lunch with Iggy Fanlo, CEO of Adbrite. He is one of the most thoughtful people I know in the online ad business and I always enjoy our conversations. He related to me how he sees ad targeting falling into four flavors:

    Geographic
    Demographic
    Contextual
    Behavioral

He noted that the first two of these, geographic and demographic, are black and white, and focused on the user. You are in one and only one location. You have one and only one gender, age or income. In each of these cases, the key is to gather a broad dataset with which to target. As a result, the largest sites and networks, with the largest datasets, will tend to be best at these flavors of targeting.

Contextual targeting is also black and white, but it is not user centric. Rather it is focused on the page. You are looking at a page that is about some topic. Search is the easiest case, where the user tells you what the page is about. Vertical ad networks with endemic advertisers are also pretty easy to contextually target because they only include sites within their desired topic. But the general case is much harder. Now the winner isn’t necessarily the one with the largest dataset of users, but rather the one with the best algorithm for figuring out what the page is about.

Behavioral targeting is not black and white, but rather shades of gray. Furthermore, it is both user centric AND page centric because behavioral targeting is the accumulated sum of historical contextual targeting. It is based not on what page you’re looking at now, but rather on what pages you’ve looked at in the past.

In this case there are advantages to both having more information about users past behavior, AND better algorithms. In it’s simplest form, retargeting, a web user who had visited Ford.com in the past will be shown Ford banner ads while on other sites. But ad fatigue limits the frequency with which one can retarget based on a single datapoint. Good behavioral targeting systems need good historical data as well as good algorithms to best manage the portfolio of advertising opportunities to a single user. Companies are using many different sources of historical data, including search history, looking for a user on the ad network, watching a user at the ISP level and even watching offline behavior.

AdBrite recently launched an Open Targeting Exchange where it will let any company with a targeting algorithm bid to be used to target ads across their network. It is a very interesting idea, and I’ll certainly be watching closely to see how it works for them.