• by Richard Frankel , August 24, 2011

Recently, there has been a spate of articles about how last-click attribution is a flawed way to measure online ad effectiveness for advertisers. Others have debated view vs. click attribution. But these debates miss a key point. The damage of misplaced attribution doesn't just fail to give credit where it's due for conversion, but also harms brands' online advertising efforts more widely -- it hampers their ability to drive upper-funnel results.

Every brand marketer knows it takes time to develop engagement with consumers.

Leading consumers through the sales funnel is a process that involves multiple advertising and marketing "touches" per person. By not measuring the real value of these touches, marketers don't get a true picture of the ROI of their online ad campaigns, and they don't maximize results against each off their marketing goals.

A customer may click on your ads multiple times before converting, each time getting a bit closer to making a purchase. Or, he may see your ads many times but never click on them -- but still get a favorable impression that leads him to remember your brand, consider it when he enters research mode, purchase your products in-store, recommend the brand or product to a friend. Though they are rarely accounted for, all of these touches can be attributed and managed as part of integrated media campaigns.

To leverage online advertising as a valuable brand-building resource, marketers must approach how they measure the impact of online ads in a totally different way. So what steps can brands take now to ditch last-click attribution and start measuring the real success of their online campaigns?

To get there, you need to break down the silos between analytics and execution. Historically, marketers have tried to measure how people interact with their brands as they travel the Web and move through the purchase funnel. But the execution of media -- the actual buying of inventory -- is often divorced from this brand analysis. Brands need to integrate analytics and execution more tightly, measuring the brand impact of every click or view, then using the findings to make optimized ad buys for every target audience at every interaction point.

By measuring and optimizing media for every touch, you leave behind the world of last click. Now you can tie together "soft" metrics, like preference and sentiment, with "hard" metrics, like sales. With the latest online tools, you can even discard the notion of hard and soft metrics, and turn all measurement into real metrics.

Melding real-time analytics and media buying is still easier said than done. Here are three steps to get you started:

Use varied methods to measure the impact of ad interaction.
There are many ways to measure the brand impact of every ad interaction, whether this interaction leads to a conversion or not. You can use real-time, in-banner surveys to measure upper-funnel metrics, like brand awareness and preference. You can use social metrics such as likes, shares, and product reviews to evaluate brand sentiment in the middle of the funnel. As you get lower in the funnel, coupon downloads are a great way to measure intent to buy. And increasing numbers of marketers can measure conversion -- that key sales event at the bottom of the funnel.

Apply data in real-time to media execution.
A relatively new advance in online marketing is using data to make real-time media-buying decisions. Marketers have always used metrics at every stage of the funnel to make decisions about what to buy and when. But only now can you turn those decision cycles from manual, annual events into real-time, automation-powered media buying. Marketers can input all the metrics they gather -- about brand impact, intent, sentiment, clicks, shares, conversion, and the like -- into the media-buying process to make the most accurate buys possible.

For the first time, digital media can actually be executed as integrated, high-impact campaigns at every point in the funnel. The silos between branding and response are being broken down as new technology is being leveraged by marketers.

Don't panic - you can start small.
By now, you're probably thinking: "I can't possibly measure 50 different touch points for each and every ad!" But even if you measure four touch points instead of one (the last click), you'll still get a more accurate view of the ROI of your campaigns. You could start by moving from measuring just last-click conversion metrics to add in a few middle-funnel analytics, such as how ad interaction drives product searches on your site, or whether your ads are shared among social networks.

Moving beyond the last click is not that difficult, especially if you take it one step at a time. For example, a major auto manufacturer we work with expanded analysis from just measuring bottom-of-the-funnel conversion (filling out a lead form) to include middle-funnel metrics like car configuration, research, and brand and model consideration.

By giving credit to consideration, the automaker actually found 20% more "hidden impact" of its lead-gen campaigns than when it just measured leads. They could actually assign a measured value to interaction with the campaign at the middle of the funnel, because it drove more activity at the bottom of the funnel. With this information, they could optimize campaigns to boost total ROI.

Multi-touch metrics can also lead to a huge uptick in real measurable sales. A laptop manufacturer that integrated multi-touch analysis across the funnel ended up doubling revenues. They found that measuring just the "last click" had resulted in massive underinvestment in media buys targeted to people in "research mode."

By putting more ads in front of these searchers, the laptop maker doubled sales. The wrong metrics and attribution don't just reward the wrong partners -- they actually cost you sales.