Measuring…no spoon required

In this week’s installment of the Marketing Speakeasy, we’re delighted to welcome Jodi McDermott, measurement whiz and senior director of product management at comScore. In considerable detail that follows, Jodi addresses some big questions related to audience measurement and web analytics.

In related news, Jodi and our own master of multi-channel metrics, Akin Arikan, are speaking at tomorrow’s OMMA Metrics conference in San Francisco. Be sure to check out their sessions if you’re attending what’s expected to be a great event.

Unica: What are the synergies from using audience measurement in conjunction with web analytics?

Jodi McDermott: Audience measurement and web analytics tools both play an important role in the digital marketer’s toolkit.  In order to understand the particular synergies between these two measurement methods, one must first understand their differences:

  • At comScore, we begin with an opt-in research panel of approximately 2 million Internet users worldwide who provide us with a thorough profile of the demographics of their household, including age, gender, household income, and others.  We use the known information about the panelist and their comprehensive clickstream traffic across the Internet, combined with census level data collection, to project a unified view of the audience visiting a particular site – we refer to this blended methodology as Unified Digital Measurement TM.
  • Web analytics tools start with site-centric census data collection and focus on allowing the user to analyze the individual requests on their site aggregated at a page, session and visitor (cookie) level for analysis. Because the data is so granular, an analyst can segment the data by campaign code, geography and URL all the way down to the hourly level if their system allows.

Audience Measurement data enables marketers to answer questions such as:

  • What is the demographic profile of visitors who come to my website?
  • How does my site benchmark against my competitive set with respect to Unique Visitors, reach, page view consumption and site duration?
  • What other sites have audiences similar to mine?

Whereas Web Analytics data enables marketers to answer questions such as:

  • How many visits occurred on my site yesterday, today or this month?
  • How are these users navigating my site?
  • Which campaigns, referring sites and search phrases drove them to my site?
  • What are the common attributes of users who register, purchase or execute some other value event on my site?

One important source of difference between the two measurements methods is the way the Unique Visitors metric is calculated and reported.  Because panel-based audience measurement methods measure at the individual user level (as opposed to the site level) it is not affected by factors such as shared usage on a machine, cookie deletion, bot traffic and users of multiple devices or browsers, when accounting for unique visitors to a site. While web analytics services also report the Unique Visitors metric, this metric can perhaps be better described as “Unique Cookies” for web analytics, because cookies and not people are what is really being counted. While it does present a census level view of behavior, it will almost necessarily overstate the number of actual unique individuals who visit a site, and this inflation factor tends to get larger the longer the period of measurement being used.

Despite the differences between these methodologies, there is also potential for synergy when the two are used in conjunction with one another. Specifically, by unifying these views of behavior marketers are better able to align high level marketing strategy with tactical site level optimization. They can devise their strategy at the top of the marketing funnel and see it through to the conversion level at the bottom of the marketing funnel to ensure that the strategy drives bottom line results.

Unica: While many companies combine audience and web analytics, the discussion on blogs or Twitter seems extremely silo’d. Why is that?

Jodi: Starting with Twitter, you only have 140 characters to get your point across :) That said, blogs and the discussions on Twitter that drive traffic to those blogs, have generally been focused around the two different areas of online measurement. This has primarily been due to the users of the data being segmented into two different silos within an organization – media/planning/research and site usability/web analytics.

Web Analytics is still a relatively small community of users and social media and discussion forums have provided a hyper-intensive environment for these conversations to take place.  This is a good thing – as Web Analysts often find themselves to be the only one in their organization who “do what they do”, social media has provided an online learning environment through sharing and testimonials that has helped to propel the industry forward.

Until recently, each measurement technique started from two different directions of data collection and analysis, one panel and the other census. These two online measurement systems have been breaking down their silos somewhat rapidly the past few years as audience measurement firms such as comScore have integrated census level data collection to better account for audiences such as work, shared usage and mobile devices. The techniques used to formulate the metrics may be different, but they are more often than not starting to use the same data sources as inputs.

Unica: Besides web analysts, who in the organization uses audience measurement tools? What can Web analysts learn from these peers?

Jodi: Audience measurement tools are used by a variety of users within an organization.  At the executive level, audience measurement provides benchmarking data on how a site is performing against its competitive universe.  For example, an online news site will use the data to determine if the online consumption of news is growing or declining as a whole.  Based on those numbers, it can also determine its share of the total audience and the demographics that comprise these audiences.

Research analysts use audience measurement data to prepare media buys for their organization or to respond to RFPs if their site sells online advertising. Since audience measurement data is produced by a third party service, media buyers trust it to provide them with a site’s demographic makeup, reach and frequency and behavioral attributes prior to executing an insertion order to run their campaign.

The more that Web Analysts understand the business questions that their internal customers are asking about who visits their website, the better they can serve them when preparing analytical decks and providing insights that will help drive the strategy of their organization.  Top level site-centric data with audience measurement research working together can provide a comprehensive picture of not just how many visitors are coming to your site, but the demographics and attributes of that audience to help you optimize your site’s business.

 

This entry was posted in Uncategorized. Bookmark the permalink.
  • http://multichannelmetrics.com Akin

    Super interview. Glad to hear that audience measurement and web analytics have been getting more aligned over time.

    While web analysts are not be the primary users of audience measurement, they will really benefit from taking a much closer look at comScore and similar solutions.

    My eyes were opened to this while I was creating web analytics wizards on my blog for troubleshooting a drop in site traffic and resulting sales. Namely, you might be searching for the problem on your own site whereas audience measurement can give you a quick idea whether your competitors are seeing similar drops of have been getting a greater share of the traffic that used to be yours.

    Thanks much to Jodi and Dan for the interview!
    Akin