Adding Behavioural Data to Survey Data by Scott Porter, Thao Duong #CASRO #MRX

Live blogging from the CASRO Digital conference in San Antonio, Texas. Any errors or bad jokes are my own.


 Adding Big Data Booster Packs to Survey Data”

Researchers may be intimidated by big data and its purported ability to replace survey data. However there are ample opportunities for the two to be combined in a powerful way. Citing case studies, this presentation illustrates how several types of business issues can benefit from such a combination. The areas featured include comparing direct-response creative concepts, better understanding of sales trends, and the drivers of observed consumer behavior. Although such analyses were possible before, they are all enhanced by the clients’ investment in capturing clean, granular data. In addition to the presenter, this paper was authored by Carlos G. Lazaro (Added Value).

Scott Porter, Vice President Methods, Added Value

  • Surveys are monthly, weekly, or data but now data is minutely and secondly and constantly
  • data is different – website visits, reviews, posts, wish lists, clicks, bookings, comments, price changes, shopping carts and it all looks different and is stored different
  • We need to have a new ad ready when the old ad tires people out. You need to be storyboarding in time to be ready for the next slot. But what if ad bombs really early, are you ready?
  • Calls and clicks are good. Ad attributes testing is good. But both together provide more.
  • Wanted to focus on people not buying enough and people who are not visiting the website enough

Behavioral Data as a Complement to Online Survey Data in Measuring Effectiveness of Online Ad Campaign”

Behavioral targeting and online survey collection are different approaches towards measuring online ad effectiveness. Traditionally, each source of data is used individually. Behavioral data has been effective in gaining tactical understanding of site traffic while survey data provides insight into the website’s impact on a brand and its ad campaign. In this paper, the author propose a methodology to combine these sources of data in order to improve market research process and increase accuracy in measurement.

Thao Duong, Senior Statistical Analyst, comScore, Inc

  • Data can be man-made via surveys or harvested via social media
  • Tested 3 ads – standard mobile banner, adhesion banner, in-text branded keyword plus control group
  • Used pixel calls as a lift measure
  • Show ads that were not available anywhere else on the internet, created just for the study
  • Most users only looked at standard banners whereas higher proportions interacted with adhesion ads and in-text branded keywords
  • Add interaction data using pixel calls to different ads provides additional information than survey data, helps you come closer to the truth, creates lift compared to standard banner ads

Other Posts

%d bloggers like this: