Insights

Metric Madness – New Challenges and Opportunities

By Steve Latham, CEO of Spur Interactive

Most savvy marketers know that the Web is a “must-have” component of every company’s advertising, marketing and communications strategy. In today’s climate, ROI and measurability are paramount. Compared to traditional media such as print, radio and television, the Web defines a new standard for measurability. This is one of the reasons why online ad spending grew more than 30% in 2006. The paradox is that those of us in the interactive marketing world are at risk of becoming victims of our own success. If we aren’t measuring the right things (and doing it correctly), we are missing the target for measuring ROI, as well as the opportunity to generate meaningful growth in sales and brand awareness through strategic online marketing programs.

The evolution of metrics
Long, long ago, in the early days of the Web (circa late ‘90s) the universal metric for measuring Web site activity was “hits” which was often mistakenly used as a proxy for visits. Then we started looking at other metrics that might help us glean insights into visitor behavior and site performance and focused on unique visitors, page views, returning visits, referring URLs and search keywords. Then things became very exciting when we started measuring conversions (e.g. sales or registrations) and tracing them back to their source (e.g. which search engine and/or keyword initiated the visit which resulted in a sale).

Armed with this data we were able to create beautiful charts with eye-popping stats that clearly showed how online ads generated clicks that converted to revenue. For many advertisers, these metrics changed the how they would view marketing and advertising going forward. Measurement became a “must-have” and ROI became the most important metric. Then we learned something new; something that would yet again require us to change the way we measured and analyzed campaign performance.

We learned that consumers were becoming smarter and more deliberate in their online research and/or shopping habits. We learned that customers like to do research before making a purchase, requesting information or taking a desired action online. We also learned that visitors might take a wide range of paths when returning to a Web site.

Given that consumers often make several visits before taking action, the Web analytics community has had to take a step back and re-evaluate what we measure and how we do it. If we are to calculate a true ROI from our advertising campaigns, we need to measure not only the most recent source of the visit, but others before it as well. Yet, most popular tracking software applications were designed to capture only the last source of the visit (referring link or URL) and measure what happens during the visit. If something good happens (e.g. a sale) it then assigns that action to the apparent source of that visit (the last link the visitor clicked before landing on the site). No credit for action is given to the ad that generated previous visits. Even worse, it’s common that the last ad clicked on (often a search ad) gets the credit for the sale. If this is how we continue to measure ROI, we’re missing the bigger picture and we may end up allocating all of our dollars to the lead source which is only valuable after awareness has already been created.

While it is no trivial task to determine what to measure, how to get the data, and how to glean insights from the data, when done right, it can provide invaluable insights into customer preferences, buying behavior and critical success factors for interactive engagement.

Case Study: Tyrell Inc.
Tyrell, Inc. is the maker of the award-winning “Zeno” skin clearing device, a revolutionary consumer instrument that uses heat to kill the bacteria that causes pimples. In scientific studies, 90% of pimples disappeared within 24 hours (to learn more visit www.MyZeno.com).

Zeno is marketed through multiple channels including thousands of physicians, high-end retailers such as Nordstrom and, as of late, every Walgreens in the United States. Like many innovative products that seem too good to be true, Zeno has generated a lot of buzz, which has translated into substantial online traffic at MyZeno.com. In addition to using the site to educate consumers and help them find a licensed practitioner (physician) near them, MyZeno.com sells Zeno directly to consumers. With the goal of optimizing sales and maximizing ROI from its advertising dollars (both online and offline), Spur Digital monitors and analyzes traffic and sales on a daily basis.

When analyzing visitor behavior and consumer engagement, we learned several interesting things:

  1. Most purchases take place on repeat visits. Of those that purchase Zeno online, only 1 in 3 purchases took place during their first visit. Most purchases take place during visits 2-6, with some sales taking place on subsequent visits.
  2. Repeat visitors often rely on search engines vs. bookmarks. One would assume that after visiting MyZeno.com, most would simply type in the URL to return to the site. The data tells a different story: 50% of those making their second visit and 40% of those making their third visit to MyZeno.com used a search engine to find their way back to the site. Search is a powerful channel, even after the initial visit.
  3. Consumers do a lot of research before making a purchase. The average visitor views 6.5 pages and stays more than 5 minutes per visit. We also found a significant percentage sign up to receive Tyrell’s newsletter. Together, these facts provided a strong indication that educational content (about Zeno, how it works, testimonials, FDA test results) is an important component of the purchasing cycle.
  4. There is a two-week lag between advertising and sales. After we learned that the majority of sales take place during visits #2-6, we analyzed the purchase data and found there was a two-week lag between the initial visit from an online ad, and the online purchase of the product. The two-week lag correlation (0.90) provided valuable insight into consumer buying cycles and is now being used for media planning and forecasting.

This case is significant for two reasons:

  1. It demonstrates the value of analyzing the right metrics. If we looked solely at clicks-to-conversions, we’d miss out on a lot of important information relating to customer behavior, sales cycles and use of online media. By understanding these issues, we can plan and forecast with greater confidence and precision.
  2. It proves that measuring session-based conversions is no longer sufficient. Given that consumers are increasingly using the Web to do homework before a considered purchase, the initial visit is not likely to produce a desired action. If a visitor originally found the site by clicking on a banner ad, and later returned to the site by doing a Google search for “zeno” before making a purchase, most tracking systems would incorrectly credit Google as the source for this sale. While Google should get the “assist” on the sale, the original source for the lead was the display ad. To properly optimize campaigns, it is critical that some credit for the result is assigned to the advertisement from which the initial visit originated.

The Tyrell case is more common than not. Using cookie-based tracking systems, we can track many of these heretofore untraceable activities. But even with sophisticated systems, reliance on cookies has its downside. Cookies are susceptible to expiring, being disallowed by browsers, and being routinely deleted. Consequently, we still don’t get a complete picture from analyzing daily or monthly site activity.

Considering the influence of the Web on offline purchases, the problems becomes even more complex. Numerous reports state that 20-25% of offline retail sales are impacted by online interaction. I believe that in some industries such as travel, electronics and other considered purchases, the real impact is much greater.

There are numerous other examples that demonstrate how pinpointing the true impact of the Web on customer awareness and intent is almost impossible. While we (the interactive marketing community) have made significant strides in quantifying the impact of online marketing efforts, we realize that even though the proverbial haystack is smaller, it’s still hard to find the golden needle.

Realizing there is no magic bullet for measuring the true impact of media on sales and brand loyalty, we must take a broader view of customer interaction. To properly measure and analyze that interaction (now referred to as Engagement, or the interaction that takes place between the advertiser and the consumer), we must expand the set of metrics we measure, and equip ourselves with the tools needed to measure them. To learn more, read Interactive Engagement and Metrics.

Steve Latham is the founder and CEO of Spur Interactive, a strategic interactive marketing agency. Steve is the President of the Houston Interactive Marketing Association and a Director of the American Advertising Federation – Houston. For more articles visit www.spurinteractive.com