Questions for Analytics-Driven Marketers

Analytics, the bedrock of interactive marketing, is fraught with complications. Some are generalized, affecting all marketers. Examples include the reported increase in the rate of Web cookie deletion and the universal problem of how to sort through a glut of data. In most cases, though, the problems of reporting and measurement are as unique to each marketer as the objectives.

To get a grasp on the dominant concerns, ClickZ polled several marketers and analytics experts on what they consider to be the biggest challenge in using Web analytics today. Each of the following respondents is a speaker at the upcoming Emetrics Summit in Santa Barbara, Calif.

Jim Sterne, President
The Web Analytics Association

The biggest challenge is making the data actionable.

This dilemma begins when an analytics tool is plugged in, turned on and starts spitting out reports, then ends with a general lack of interest and finally dies in a fog of ennui.

The first round of Web analytics reports are interesting. They can immediately highlight anomalies that point out specific problems. These are the low hanging fruit, the impediments that require a minimum of intervention yet yield a maximum of value: the broken link nobody knew about, the “Buy Now” button below the fold, the content page with no call-to-action. Quick fixes, great results.

But the next round of reports do not contain such obvious pointers and now the heavy lifting begins. The data must be sorted and sifted, sliced and diced, and generally teased into revealing something actionable — some golden bit of insight. It’s very hard work.

Sam Decker
Senior Manager, Marketing & E-business
Enterprise Transactional Group
Dell

The big opportunity for Web analytics for Dell is not narrowly defined to Web measurement. The consumption of Web analytics has evolved from the Web team, to general marketers, to other partners who are interested in customer insights. As such, we have become much more broad in how we think about Web analytics.

True customer and business understanding does not come from online metrics alone — they come from a combination of online data, offline data, order data, financial data, and external data. The ability to integrate these standardized data sets, analyze quickly, and communicate insights in a meaningful way is a vision we’re working towards. In the end, it’s all about presenting customer driven information and responding to customer communicated needs.

Jim Novo, Consultant
The Drilling Down Project

Biggest challenge: the lack of an analytical culture in many companies.

Marketers driven by Web analytics are very accountable for their results. If they work in a company where there isn’t this level of accountability for other marketing activities, they may find it difficult to “fit in” with that low accountability marketing culture and even catch heat from the people in it.

Analytics bring accountability, and accountability breeds fear. Many marketing people feel threatened when someone shows up and says, “Hey, we’re going to be able to prove whether the Web marketing stuff you do actually works. Isn’t that neat?”

The analytical culture is different from the general business culture at most companies. In the high accountability business model, people are driven by testing and performance-based results. Failure is embraced as a learning experience, and continuous improvement cycles are a way of life. This is a highly proactive, aggressive business model, and the culture has unique needs.

If the company is embracing analytics broadly, employees need to be briefed on what to expect and how the analytical culture operates. If this situation isn’t properly handled, often what you get is the opposite of what should happen; instead of increased productivity, people don’t participate and the business suffers from a lack of creative thought. Much of what is called “CRM failure” is due to this issue.

Avinash Kaushik
Senior Manager, Web Research & Analytics
Intuit

The Web analytics challenge at Intuit is two-fold: one, how to focus attention on “outcomes” for the company from Web site interactions; and two, how to ensure that the voice of the customer (VOC) has equal, if not more, representation in decision making as internal company needs.

Most Web Analytics solutions provide such an overwhelming array of reports that most users cannot focus on the key metrics that drive successful outcomes. Rather than plot graphs showing trends of traffic or page views or directories viewed or browser versions or traffic peaks during the day, it is important to focus on what the outcome was of that Web interaction and if it was a desired outcome. Our approach to solving this is to build a centralized data warehouse that collects aggregated Web behavior metrics and ties that with very rich outcomes data and the key metadata for the company. This empowers sophisticated decision making about not only the “hows” and “whys” of the outcomes; it also allows drill-down and drill-around for a massive pool of data. Coupled with an uncluttered, powerful, clickstream analysis tool like ClickTracks, this approach has helped Intuit elevate our sophistication in “Web analytics.”

We all think we know what customers want, yet it almost always turns out we are not on the money… even though almost all of us in the industry have pop-up surveys on our sites. At Intuit, we have adopted the ACSI (American Customer Satisfaction Index) methodology to infuse a statistical rigor to understanding customer experience, from their perspectives. We can measure customer satisfaction on the site while measuring correlations between that customer satisfaction score and various key site elements. (Think navigation, search, content, etc.) The methodology, from ForeSee Results, also predicts key future behavior. (Think likelihood to return or buy.) But perhaps the lynchpin in this approach is the raw VOC responses that we get from the open-ended responses. The statistical approach isolates key areas that need attention, and then customer VOC is an incredible source of potential solutions.

Matthew Berk
Founder and Chief Technology Officer
local-I

At Open List, we face two major challenges to making the most of our Web analytics tools. The first is keeping pace with change. Within the past nine months, our site, which is a local search engine, has grown dramatically, from restaurant search in one local market, to coverage of close to a million restaurants [and] hotels. While the page tags are easy enough to change, major shifts in site architecture… have implied equally tricky changes to our reporting tools and how we read them.

The second challenge, which is exacerbated by the first, is finding the time and resources required to maintain and learn from our analytics tools. Although the data we have is precious in terms of helping us understand visitor search behavior so that we can optimize our algorithms, finding the manpower to conduct the analysis is difficult. Because we’re a startup, we have few heads, all of whom are already overtaxed; this means we only get to do the analysis and optimization on a monthly basis. Had I my druthers, we’d use the tools far more often for this purpose. The counterintuitive problem with having great analytics tools — we use Omniture — is allocating enough bandwidth to use them to optimize the site. My recommendation to any site considering getting a decent toolset is to make sure at least one employee is responsible — on a dedicated basis — for leveraging site analytic data to improve the site.

Related reading

/IMG/853/275853/gmail-logo-2013-320x198
activist
facebook-organic-reach
YouTube-logo-full_color
<