Imagine this: You start a new job overseeing the Internet channel at a traditional retailer. Your first priority is to dig into the data to learn how your customers are interacting with the Web site. You meet with the vice president of research and ask for the latest reports. The response is shocking: “Customer data? We don’t have any customer data — none that’s recent anyway. We haven’t done a customer survey in over nine months, and we don’t have the budget to do one now, but put it in your budget request for next year.”
You explain that self-reported data (e.g., what customers tell you through a customer survey) wasn’t exactly the kind of information you were looking for, although it is nice to have. What you’re really after is observational data (e.g., what customers do on the Web site). The vice president of research looks at you inquisitively and asks, “You mean the daily sales report? You should have gotten one this morning. It recaps all the sales transactions over the previous 24 hours. Total number of transactions, number of purchasers, average transactions per purchaser, average purchase amount, typical stuff. That should give you the information you need.”
You take a deep breath and struggle to maintain your composure. Transaction data is good to have, but it isn’t what you’re after either. No, you’re looking for reports that compile information from the Web logs — reports that provide detailed information on how consumers interact with the Web site. You want to know how consumers navigate the site, where consumers come from, how many times consumers visit before making a purchase, and so on. The vice president of research replies, “We don’t have that. It’s in the Web logs, but what good is that information anyway? We really just need a new customer survey. Put it in your budget request for next year. Just use the daily sales reports until then.”
You realize that you’ve just uncovered the first ghost at the new company. “Ghosts” are the things that would have scared you away from the job if you knew they existed — things conveniently tucked in a closet during the interview process. Or maybe you just didn’t look hard enough. Either way, the ghost is now out in the open and needs to be addressed. Your new company fails to recognize the value of analyzing observed customer data. Your job is to become an evangelist, convincing those around you and above you that analyzing observed customer data will pay significant dividends. That old job you left isn’t looking so bad anymore, is it?
Take a deep breath. All is not lost. The customer data required to start your analysis exists in the Web logs. Many companies, yours included, fail to recognize the many uses of observed customer data stored in Web logs. A few examples lie below:
- Advertising optimization. Where were your site visitors before landing on your Web site (banner ads, emails, search engines)? Although this sounds rather simple, the number of companies that fail to measure the effectiveness of their advertising vehicles, much less compare the effectiveness of multiple advertising vehicles against one another, is amazing. When tied to purchase data, this information allows you to target your advertising dollars more effectively by identifying the advertising vehicles that yield the best customers.
- Point of entry. Which Web pages do site visitors first land on most frequently (the home page or pages deep within the Web site)? It’s not intuitive, but site visitors may not access your Web site through the home page. Search engines and bookmarks are the most common vehicles leading consumers to pages deep within your Web site. Identifying popular, alternative entry points helps you identify the areas of the site consumers find most valuable.
- Navigation optimization. How do site visitors navigate the Web site (search box, links on the rail, links on the masthead)? This is analogous to understanding the traffic patterns in a physical store. There’s a reason milk is at the back of a supermarket — it’s called merchandising. Understanding consumer’s natural traffic patterns on the Web site provides valuable data that can be used to more effectively merchandise the Web site. This also may help you identify bottlenecks that are frustrating consumers and leading them to flee your Web site.
- Visitor-to-purchaser frequency. How many times does the average consumer visit the site before making her first purchase? How many times does she visit before making a second purchase? How many times does she visit before making a third purchase? Those who have worked with me are fond of this conversation. Identifying key metrics for acquisition, retention, and stimulation and employing marketing programs to positively impact these metrics are fundamental to maximizing the lifetime value of a customer.
So much data is contained in the Web logs that it can overwhelm even the most experienced analyst. The only way to eat this elephant is one bite at a time. Identify a few key pieces of knowledge that, if known, would lead to tactical recommendations to improve the business performance. Make sure to pick the low-hanging fruit, the areas that tend to yield the greatest results with the least amount of effort. Then request that the necessary observed data be retrieved from the Web logs. And don’t worry — you won’t need fancy or expensive software to perform the basic analysis needed to get started.
Now, get back to work. You have a lot of data to analyze and can’t afford to be surfing the Internet during business hours at your new job.
Marketers need to know what’s in their data and trim out the filler to provide continuous, data-driven ROI for their brands.
A new starter in Team SaleCycle recently asked me the following question… “Wouldn't they just come back anyway?”
American Apparel's chief digital officer discussed the future of retail, the importance of delivering value to the consumer, and strategies for an IoT and omnichannel world.
Every marketer has been sitting with his or her analytics team, reviewing an overwhelming spreadsheet of data points. It tends to hurt ... read more