Adding Behavior Metrics to Marketing ROI Measurement
Luth Research Whitepaper
Adding Behavior Metrics to Marketing ROI Measurement
Prepared By: Becky Wu, Ph.D. Vice President, Research Luth Research (619) 234-‐5884 ext. 8053 [email protected]
Table of Contents
WHY BEHAVIOR DATA ARE A GAME CHANGER FOR ROI MEASUREMENT AND RESEARCH IN GENERAL ........................ 2 ZQ DIGITAL TRIBE™: HOLISTIC MEASUREMENT THROUGH SURVEY AND WEB BEHAVIORAL DATA .............................. 3 TRADITIONAL SURVEY METRICS YIELDED MEDIOCRE RESULTS IN ROI .................................................................... 4 WEB BEHAVIORAL DATA DELIVERED UNFORESEEN INSIGHTS IN A BIG WAY ............................................................ 5 CAPTURING THE BENEFITS AFFORDED BY BEHAVIORAL DATA ............................................................................... 6
Why Behavior Data are a Game Changer for ROI Measurement and Research in General The need for measuring marketing accountability is more pressing than ever before. Marketers are both blessed and challenged with a plethora of new and evolving media and marketing technologies. The fast pace and increasingly complex nature of marketing campaigns lead to a majority of marketers rushing new marketing campaigns in to implementation based on intuition, with only 13% of marketers citing any form of quantitative assessment prior to launch (MarketingProfs.com Benchmark Study, 2005). The challenge with measuring return on investment (ROI) lies in not only determining the appropriate metrics to focus on, but also having access to the right type of market data. As marketing practices and technologies continue to evolve, research measurement and methodology for ROI are due for an update. Inclusion of behavior data is one promising avenue to enhance ROI measurement. Why do behaviors matter? In the context of market research, behavior data matter for at least two important reasons. First, behavioral data provide a more truthful understanding of what is happening in the consumer’s life. The past decades of survey research are primarily based on respondents’ recall and self reporting. The accuracy we can gain by having actual behavioral data is an obvious benefit. Second, like consumer attitudes, consumer behavior is an essential part of the human decision process. More often than not, behavior is one of the direct outcomes that marketing aims to influence or change. For example, is your marketing campaign making any meaningful impact on how your customers interact with your brands online, in the form of social content sharing, or performing less comparative shopping? Being able to capture this type of consumer behaviors is a pre-‐condition to measuring the ROI of your campaign. This aspect of market intelligence is especially more important given the common consensus that we are not always able to articulate or explain what we do. However, there has not been a feasible and financially viable way to measure broad behaviors until the advent of Internet-‐based tracking technologies in recent years. Even when these tracking technologies become available, they have been primarily used by big data players such as Google and Yahoo to benefit their advertising operations. Separately, there are companies like Adobe/Omniture garnering a wide range of Internet data with a focus to provide syndicated insights, which are not necessarily tailored to each brand or business. In short, the vast majority of businesses do not have access to cost effective digital behavior data that are gathered, analyzed and delivered based on their own business requirements. With the development of ZQ Intelligence™ at Luth Research, we now have an opportunity to change the status quo. ZQ Intelligence is a web-‐based behavior tracking technology that allows businesses to step inside consumers’ and customers’ digital lives with transparency. With their explicit permission, individuals and respondents download this technology onto their computers, providing access to their online activities on a continuous basis. With this unprecedented research capability clients gain deep, direct visibility into the daily digital lives of their customers and prospects. The case study in this whitepaper illustrates the compelling value of digital behavioral data by comparing insights gained by using survey data alone with those from an integrated approach where behavioral data has been added into ROI measure, which otherwise would not have been revealed. Luth Research, LLC.
2
5/7/2012
ZQ Digital Tribe™: Holistic Measurement through Survey and Web Behavioral Data As part of the background for the case study, Luth Research was tasked by a digital publisher in the health and wellness information space to measure ROI for one of its advertisers. This publisher operates with a business model under which it serves as a content curator on behalf of consumers seeking relevant wellness knowledge and information. It invites leading brands in healthcare products and services to be knowledge experts providing first-‐hand information on select topics as well as to directly interact with consumers in need of assistance, guidance or information on its website. The ultimate ROI question from these leading brands, who are the unconventional advertisers in this case, is whether or not consumers come away from this experience with a more positive brand perception, a higher level of brand awareness, and/or a stronger lift in brand and product interest. The publisher client intended to leverage the holistic approach combining both survey and behavioral data offered by ZQ Intelligence to assess the ROI for a dental product brand. To address the above goals, Luth Research implemented a 60-‐day ZQ Digital Tribe with a total of 100 target consumers. A ZQ Digital Tribe provides clients an easy access to a defined group of respondents or consumers who agree to allow their online behaviors to be tracked for a designated period of time. For this study, all target consumers were recruited from Luth Research’s online research panel and had to meet the criteria of having near-‐term interest or on-‐going interest in dental health or specific dental issues. The Tribe was set up with an experimental design. The 100 respondents were randomly assigned into two groups with approximately 50 people in each group: a control group and a test group. Both groups went through an identical process in the first 30 days of web behavioral tracking. They all completed the first survey within the first 30 days, which measured the baseline brand awareness, perception and other key metrics. The initial set of web behavioral data between the control group and the test group were collected and analyzed to ensure there were no significant differences between these two groups of respondents in terms of web browsing. Upon completing the first 30 days of tracking, the test group was sent an email notification, providing a brief introduction about the publisher and a request for the respondents to visit the publisher’s website at least once a week for the next 30 days. Respondents could visit the website as much as they wished to above the aforementioned minimum requirement. Once the introduction and requirement was communicated to the test group, respondents in both groups were tracked for another 30 days At the end of the entire 60-‐day tracking, all respondents from both the test and control groups completed a final survey measuring post-‐tracking brand awareness, brand perceptions, and other relevant metrics. Both survey data and web behavioral data were integrated into the final analysis. Only web behavioral data from the second 30 days of tracking were used to determine the differences in behaviors between the control group and the test group. Luth Research, LLC.
3
5/7/2012
Traditional Survey Metrics Yielded Mediocre Results in ROI Following the traditional practice in ROI measurement, we first examined the differences in brand awareness and brand perceptions between the control group and the test group. Both metrics were survey based metrics. As shown in Figure 1.0. the test group posted a small 3 to 6 points of increase in brand awareness towards to the dental brand of interest across three product categories. Figure 1.0.
In addition, the surveys measured a total of seven brand perception attributes for the brand. The results showed increases in the test group’s selection on four attributes. Compared to the control group, the test group had a 7 to 10 point lead in the likelihood of including the four descriptors in their perception of the target brand. Figure 2.0.
Luth Research, LLC.
4
5/7/2012
If we were to focus on awareness and perceptions and stop here, the lifts from exposure to and experience with the publisher’s website as demonstrated by the test group’s data were not impressive.
Web Behavioral Data Delivered Unforeseen Insights in a Big Way Next, we turned to the web behavioral data to explore if there were any meaningful differences between the control group and the test group as a result of the latter group’s exposure to the publisher’s website. The first metric examined was relevant website visitation, which was defined as visits to two types of websites: 1) dental health related websites including all websites sponsored by or belonging to the target dental brand, and 2) general health or wellness websites, on which the target dental brand is actively present. When comparing the control group and the test group (see Figure 3.0.), the contrasts were crystal clear. The test group exhibited a 38% lift in the volume of visits to dental health related websites, and a 58% lift for general wellness websites during the second 30 days of tracking after respondents in this group had the experience with the publisher’s website. This data point provided solid confirmation on the ability of the publisher to drive relevant traffic to websites where the dental product brand advertiser maintains a strong presence and have another opportunity to reinforce brand messages. Figure 3.0.
Following the analysis of relevant website visitation, we further looked at the volume of relevant searches as another key behavioral data point. The findings, as shown in Figure 4.0., proved to be another evident indicator of the positive impact of the publisher’s website on consumers’ interest towards the dental product brand. During the 30 days of tracking, the test group conducted searches Luth Research, LLC.
5
5/7/2012
related to the advertiser as well as dental health ten times more often than did the control group. When actively searching for the advertiser brand and product, consumers are clearly expressing an interest in engaging with the brand. Figure 4.0.
Taking the behavioral data a step further, we drilled down on the individual elements and areas of content on the publisher’s website. The goal was to determine which specific part of the website plays the most significant role in driving relevant traffic. The data revealed that the “Topics” area of the website has the biggest contributing influence in leading visitors to other websites relevant to the dental product brand after they had left the publisher. This type of insights add value to the ROI measurement in that it tells the client what works particularly well and what they should do more of.
Capturing the Benefits Afforded by Behavioral Data Drawing from the above case study findings, we see significant opportunities in integrating behavioral data into the mix of measurements for ROI. By only focusing on traditional survey based metrics such as brand awareness and perceptions, researchers may be missing the real “A-‐ha” in consumers’ experience with brands, which are best manifested in their behaviors towards the brands. Behavioral data present a level of accountability that is often lacking in survey data. This accountability is important for researchers and marketers to accurately assess the success of the target marketing practices.
Luth Research, LLC.
6
5/7/2012
Today’s technologies afford us the tools to access behavioral data more easily than ever before. For researchers and marketers seeking better ways to get to true knowledge, adding behavioral metrics into ROI measurement is a viable, effective way to deliver on newer insights.
Luth Research, LLC.
7
5/7/2012
Adding Behavior Metrics to Marketing ROI Measurement
Prepared By: Becky Wu, Ph.D. Vice President, Research Luth Research (619) 234-‐5884 ext. 8053 [email protected]
Table of Contents
WHY BEHAVIOR DATA ARE A GAME CHANGER FOR ROI MEASUREMENT AND RESEARCH IN GENERAL ........................ 2 ZQ DIGITAL TRIBE™: HOLISTIC MEASUREMENT THROUGH SURVEY AND WEB BEHAVIORAL DATA .............................. 3 TRADITIONAL SURVEY METRICS YIELDED MEDIOCRE RESULTS IN ROI .................................................................... 4 WEB BEHAVIORAL DATA DELIVERED UNFORESEEN INSIGHTS IN A BIG WAY ............................................................ 5 CAPTURING THE BENEFITS AFFORDED BY BEHAVIORAL DATA ............................................................................... 6
Why Behavior Data are a Game Changer for ROI Measurement and Research in General The need for measuring marketing accountability is more pressing than ever before. Marketers are both blessed and challenged with a plethora of new and evolving media and marketing technologies. The fast pace and increasingly complex nature of marketing campaigns lead to a majority of marketers rushing new marketing campaigns in to implementation based on intuition, with only 13% of marketers citing any form of quantitative assessment prior to launch (MarketingProfs.com Benchmark Study, 2005). The challenge with measuring return on investment (ROI) lies in not only determining the appropriate metrics to focus on, but also having access to the right type of market data. As marketing practices and technologies continue to evolve, research measurement and methodology for ROI are due for an update. Inclusion of behavior data is one promising avenue to enhance ROI measurement. Why do behaviors matter? In the context of market research, behavior data matter for at least two important reasons. First, behavioral data provide a more truthful understanding of what is happening in the consumer’s life. The past decades of survey research are primarily based on respondents’ recall and self reporting. The accuracy we can gain by having actual behavioral data is an obvious benefit. Second, like consumer attitudes, consumer behavior is an essential part of the human decision process. More often than not, behavior is one of the direct outcomes that marketing aims to influence or change. For example, is your marketing campaign making any meaningful impact on how your customers interact with your brands online, in the form of social content sharing, or performing less comparative shopping? Being able to capture this type of consumer behaviors is a pre-‐condition to measuring the ROI of your campaign. This aspect of market intelligence is especially more important given the common consensus that we are not always able to articulate or explain what we do. However, there has not been a feasible and financially viable way to measure broad behaviors until the advent of Internet-‐based tracking technologies in recent years. Even when these tracking technologies become available, they have been primarily used by big data players such as Google and Yahoo to benefit their advertising operations. Separately, there are companies like Adobe/Omniture garnering a wide range of Internet data with a focus to provide syndicated insights, which are not necessarily tailored to each brand or business. In short, the vast majority of businesses do not have access to cost effective digital behavior data that are gathered, analyzed and delivered based on their own business requirements. With the development of ZQ Intelligence™ at Luth Research, we now have an opportunity to change the status quo. ZQ Intelligence is a web-‐based behavior tracking technology that allows businesses to step inside consumers’ and customers’ digital lives with transparency. With their explicit permission, individuals and respondents download this technology onto their computers, providing access to their online activities on a continuous basis. With this unprecedented research capability clients gain deep, direct visibility into the daily digital lives of their customers and prospects. The case study in this whitepaper illustrates the compelling value of digital behavioral data by comparing insights gained by using survey data alone with those from an integrated approach where behavioral data has been added into ROI measure, which otherwise would not have been revealed. Luth Research, LLC.
2
5/7/2012
ZQ Digital Tribe™: Holistic Measurement through Survey and Web Behavioral Data As part of the background for the case study, Luth Research was tasked by a digital publisher in the health and wellness information space to measure ROI for one of its advertisers. This publisher operates with a business model under which it serves as a content curator on behalf of consumers seeking relevant wellness knowledge and information. It invites leading brands in healthcare products and services to be knowledge experts providing first-‐hand information on select topics as well as to directly interact with consumers in need of assistance, guidance or information on its website. The ultimate ROI question from these leading brands, who are the unconventional advertisers in this case, is whether or not consumers come away from this experience with a more positive brand perception, a higher level of brand awareness, and/or a stronger lift in brand and product interest. The publisher client intended to leverage the holistic approach combining both survey and behavioral data offered by ZQ Intelligence to assess the ROI for a dental product brand. To address the above goals, Luth Research implemented a 60-‐day ZQ Digital Tribe with a total of 100 target consumers. A ZQ Digital Tribe provides clients an easy access to a defined group of respondents or consumers who agree to allow their online behaviors to be tracked for a designated period of time. For this study, all target consumers were recruited from Luth Research’s online research panel and had to meet the criteria of having near-‐term interest or on-‐going interest in dental health or specific dental issues. The Tribe was set up with an experimental design. The 100 respondents were randomly assigned into two groups with approximately 50 people in each group: a control group and a test group. Both groups went through an identical process in the first 30 days of web behavioral tracking. They all completed the first survey within the first 30 days, which measured the baseline brand awareness, perception and other key metrics. The initial set of web behavioral data between the control group and the test group were collected and analyzed to ensure there were no significant differences between these two groups of respondents in terms of web browsing. Upon completing the first 30 days of tracking, the test group was sent an email notification, providing a brief introduction about the publisher and a request for the respondents to visit the publisher’s website at least once a week for the next 30 days. Respondents could visit the website as much as they wished to above the aforementioned minimum requirement. Once the introduction and requirement was communicated to the test group, respondents in both groups were tracked for another 30 days At the end of the entire 60-‐day tracking, all respondents from both the test and control groups completed a final survey measuring post-‐tracking brand awareness, brand perceptions, and other relevant metrics. Both survey data and web behavioral data were integrated into the final analysis. Only web behavioral data from the second 30 days of tracking were used to determine the differences in behaviors between the control group and the test group. Luth Research, LLC.
3
5/7/2012
Traditional Survey Metrics Yielded Mediocre Results in ROI Following the traditional practice in ROI measurement, we first examined the differences in brand awareness and brand perceptions between the control group and the test group. Both metrics were survey based metrics. As shown in Figure 1.0. the test group posted a small 3 to 6 points of increase in brand awareness towards to the dental brand of interest across three product categories. Figure 1.0.
In addition, the surveys measured a total of seven brand perception attributes for the brand. The results showed increases in the test group’s selection on four attributes. Compared to the control group, the test group had a 7 to 10 point lead in the likelihood of including the four descriptors in their perception of the target brand. Figure 2.0.
Luth Research, LLC.
4
5/7/2012
If we were to focus on awareness and perceptions and stop here, the lifts from exposure to and experience with the publisher’s website as demonstrated by the test group’s data were not impressive.
Web Behavioral Data Delivered Unforeseen Insights in a Big Way Next, we turned to the web behavioral data to explore if there were any meaningful differences between the control group and the test group as a result of the latter group’s exposure to the publisher’s website. The first metric examined was relevant website visitation, which was defined as visits to two types of websites: 1) dental health related websites including all websites sponsored by or belonging to the target dental brand, and 2) general health or wellness websites, on which the target dental brand is actively present. When comparing the control group and the test group (see Figure 3.0.), the contrasts were crystal clear. The test group exhibited a 38% lift in the volume of visits to dental health related websites, and a 58% lift for general wellness websites during the second 30 days of tracking after respondents in this group had the experience with the publisher’s website. This data point provided solid confirmation on the ability of the publisher to drive relevant traffic to websites where the dental product brand advertiser maintains a strong presence and have another opportunity to reinforce brand messages. Figure 3.0.
Following the analysis of relevant website visitation, we further looked at the volume of relevant searches as another key behavioral data point. The findings, as shown in Figure 4.0., proved to be another evident indicator of the positive impact of the publisher’s website on consumers’ interest towards the dental product brand. During the 30 days of tracking, the test group conducted searches Luth Research, LLC.
5
5/7/2012
related to the advertiser as well as dental health ten times more often than did the control group. When actively searching for the advertiser brand and product, consumers are clearly expressing an interest in engaging with the brand. Figure 4.0.
Taking the behavioral data a step further, we drilled down on the individual elements and areas of content on the publisher’s website. The goal was to determine which specific part of the website plays the most significant role in driving relevant traffic. The data revealed that the “Topics” area of the website has the biggest contributing influence in leading visitors to other websites relevant to the dental product brand after they had left the publisher. This type of insights add value to the ROI measurement in that it tells the client what works particularly well and what they should do more of.
Capturing the Benefits Afforded by Behavioral Data Drawing from the above case study findings, we see significant opportunities in integrating behavioral data into the mix of measurements for ROI. By only focusing on traditional survey based metrics such as brand awareness and perceptions, researchers may be missing the real “A-‐ha” in consumers’ experience with brands, which are best manifested in their behaviors towards the brands. Behavioral data present a level of accountability that is often lacking in survey data. This accountability is important for researchers and marketers to accurately assess the success of the target marketing practices.
Luth Research, LLC.
6
5/7/2012
Today’s technologies afford us the tools to access behavioral data more easily than ever before. For researchers and marketers seeking better ways to get to true knowledge, adding behavioral metrics into ROI measurement is a viable, effective way to deliver on newer insights.
Luth Research, LLC.
7
5/7/2012
