Embracing Real-Time Attribution Models

Like click-through metrics, I’ve never been a fan of most traditional attribution metrics.

Just the idea that a single consumer event like seeing a TV spot, hearing a radio ad, or clicking an online banner is 100 percent responsible for a consumer’s actions doesn’t ring true.

Most of us have a pattern of behavioral stages that lies between us and the decisions we make. We evaluate the personal value of a new product or service, based on whether or not it fits our needs, before we make a purchase. The AIDA (Awareness, Interest, Decision, Action) funnel model, while not very in-depth, still does a good job of identifying multiple stages of consumer behaviors when it comes to making a buying decision.

As a result, it makes very little sense to think that the behavior of an individual consumer, who has had a lifetime of branding messages regarding Ford trucks, is suddenly going to make a dash to a local dealership because of the influence of a single ad seen during a football game.

However, the challenge that most marketers have faced in the past is how to accurately identify and evaluate each of the touchpoints that a brand and a consumer make along a purchase path in order to determine how to divide the credit given for an acquisition accurately. The idea of “full-path to conversion” analysis has been a dream for marketers for years, but the challenge has always been how to pull it off.

I was recently pointed toward Stratos Analytics by a friend because of the in-roads they’re making toward the full-path analysis challenge. While Stratos is currently focused on helping rich media advertisers do a better job of reigning in the dataflow for campaign optimization purposes, it’s also starting to roll out attribution tracking features that allow advertisers to track both online and offline campaign touchpoints.

According to Steve Terrell, CEO of Stratos, it’s important to accurately identify the specific relationships between consumer behaviors and different media because every point of interaction tells a different story.

“The patterns we are able to identify allow us to see things as granular as how different advertisers take ownership of different locations on websites,” Terrell says. “This data not only informs advertisers about where they should be running their ads but also allows them to quickly optimize ongoing campaigns.”

Cade Herzog, principal at Dallas-area Online Performance Marketing (OPM) and a Stratos Analytics customer, agrees. “People tend to connect with a brand through several different touch points. As a result, the time between these points of contact can be greater than the life span of a cookie. Because Stratos is using more persistent Flash temporary files for tracking purposes we are able to gain a better view of the significance of each touch.”

Herzog says that OPM was able to get a 200 percent increase in actionable data (while running its cookie-based campaign tracking in tandem) by using the Stratos Analytics solution to allow the company to reduce the potential understatement of total visits. This approach has also allowed the firm to start to identify “look alike” attribution patterns that tie into meeting advertisers’ audience’s targeting needs.

“Again, this also means that we can start to identify what pathways or conditions lead to conversion points/sales events based on the ways that consumers interacted with the different campaign elements. This means that we can now give advertisers actionable guidelines for campaign optimization based on this data.”

Herzog said that OPM is doing a number of cross-media studies in-house and has seen huge percentage increases for campaigns that use multiple media channels to get the word out. “It seems that an advertiser’s credibility comes from being seen in different channels. Online display ads only may gain a little attention but display ads combined with search boosts a brand’s credibility in the eyes of consumers.”

While I applaud the new tracking solutions that companies like Stratos Analytics are rolling out, I think it’s also important to stress that the solid fundamentals of advertising have never been more important. Advertisers still need to set comprehensive goals and aim for them. Yes, measure everything, but still rely on human insight and good planning (and optimization) to identify what’s working.