Some marketers get a warm and fuzzy feeling every time they think of how well Web analytics serve their companies. That's especially true for Carol Ott, director of e-commerce finance and Web analytics for San Diego-based PETCO.
Ott explains how marketers should pay far more attention to Web site traffic than first click and last click—there's plenty of information in between, too. Target Marketing spoke with Ott after her Dec. 10 session at NCDM in Orlando, Fla. titled "Customizing Clicks: Using Customer Data for Optimized Campaigns."
Target Marketing: Based on your experience, how should direct marketers analyze site behavior for better marketing attribution?
Carol Ott: I think the best way to look at marketing attribution is to look at first click, look at last click and also doing a mix, because many people come in for four to seven visits to your Web site. PETCO is about serving real people that come to our stores on a regular and loyal basis. Knowing how they interact with us over time and through different channels is imperative if we are going to know when and where to present them with products and offers that will meet their needs.
Looking only at a single visit in a store or on the site is really short-sighted for our goals. For example, if you don't look at first click, you could be removing a marketing source that's introducing first-time visitors to your brand. And if you don't hit on the middle points, then you could be missing an opportunity to remind them about your brand. In my experience, it is really important to look at all touchpoints and not rely on a simple first- or last-click model.
TM: How should companies measure marketing ROI based on a complete picture of customer behavior?
CO: For a true, complete picture analysis, you need to bring in all the channels customer[s] touch. You need to build a complete picture of your visitors to know how they interact with stores, [the] call center and your Web site. If your company has a loyalty program, this should also be brought into the mix. We're in the process of changing the user experience on the Web based on their information on their loyalty program. If they're always an in-store shopper, it'll promote all the store promotions; if they're always an online shopper, it'll be a much different experience. That's our goal in the future; we're just not there yet.
TM: How should direct marketers gain insight into the importance of cross-sell offerings?
CO: We're actually doing a lot of testing with cross-sell offerings across the site from the homepages through the cart. Companies should look at finding out how people are browsing and buying. It is important to tweak the algorithm to serve up the best possible recommendations. It's definitely highly beneficial in increasing your average order value.
TM: PETCO already provides customers with a lot of cross-sell opportunities, correct? Why continue to test?
CO: Yes, but we're still testing. In this economy, it's really important to justify marketing investments based on results and to adjust programs along the way. Your product assortment is continually changing. What a person purchases together one month may differ in the future. There are definitely cycles for some products and product offerings. Everything's always an ongoing test. Yes, we have a lift now, but how can we make that lift better? And so we're always improving upon it.
TM: What value do promotions have, and how do they work together online to optimize clicks?
CO: You have very limited space on your Web site and very limited time to engage visitors before they abandon your site. It is important to utilize the space to engage someone to clickthrough to see the actual offering and/or to make a purchase.
This is where benchmarking comes in—you want to spot trends and compare your performance against your competitors, or even against how you were doing last year. We've created benchmarks on our placements, and we're getting to a point where we know our dollars per day that we should be seeing in those placements. We focus on many key performance indicators to determine successfulness: dollars per day, sales of advertised products and shift in product views (during [the promotion] and prepromotion).