Building a better online media plan

Consumer packaged goods marketers have been slow to incorporate the Internet into their marketing and media plans. Some of the questions marketers are asking themselves are

• How can I find and target consumers on the Internet?
• Can online media drive offline sales of consumer packaged goods?
• If online media can drive offline sales, how can I measure the return on this media investment?

The following article addresses how online marketing can play a key role in your media mix strategy. Traditional Media and the InternetTraditional media, such as print and TV advertising, has eluded the scrutiny placed on trade and consumer promotions. Even with the extensive fragmentation that has taken place, many media planners are still using age and gender demographics to determine media placements and spending levels. For anyone who has been in the business for a while, “Women Age 22–49 with 2+ Children” should sound familiar. These demographics are certainly valuable, but for many marketing efforts, other consumer identification methods can also be utilized efficiently and effectively.

There are a number of critical media issues that marketers are grappling with. Consumers have many product choices, media channels and shopping channels. Too much media is wasted. Marketers want to target consumers more efficiently through accountable, relevant measurement of campaigns. Marketers are also looking to find their most valuable customers, both offline and online. The Internet has had a huge impact on consumer behavior. The Internet provides consumers with a completely new way of accessing information and shopping. Today, consumers are online more than ever before, and the penetration rates will continue to grow [See chart 1].

The Internet now represents 11% of media consumption, a significant measure compared to newspapers (5%) and magazines (3%). Even when compared to television (50%) and radio (31%), the Internet should be considered a major force. And Internet media consumption is expected to increase +30.2% by 2005, while TV and radio expect single-digit increases, and newspapers and consumer magazines are forecast to decrease in consumption.

Targeting Consumers
Ideally, marketers would like to deliver their message to just those consumers who are identified as “high potential.” The Internet provides the opportunity to deliver these “direct-to-consumer” marketing programs and establish an ongoing dialogue with key consumer groups to reinforce brand and/or store loyalty. Combined with offline diagnostic tools such as consumer panel data and other databases like frequent shopper data, the Internet can be used to identify consumers, communicate to them and evaluate marketing programs, whether they are executed online, offline or in-store.

The first step is to establish a goal for the campaign—increased sales, increased loyalty, etc. The framework for direct-to-consumer marketing is to first identify key consumer groups by creating a profile of the consumer from the integration of online and offline purchase data captured by longitudinal consumer panel data and then score that consumer profile against larger databases such as a large Internet portal like Yahoo!. The second step is to run a specific online campaign aimed at the target audience. The third is to evaluate the campaign by analyzing the purchase behavior of the profile group that was exposed to the campaign versus the group that was not exposed—a true ROI measurement.

The graph below shows the spectrum of targeting that is available to marketers today, starting with untargeted “mass reach” programs and extending to statistical applications that will enable more efficient/effective direct-to-consumer marketing [See chart 2].

The traditional methods of targeting include demographic and geographic models that deliver consumer profiles based on age, gender and DMA. These provide information on ZIP code, income and the number of children in the household.

The combination of actual purchase behavior via a consumer panel and an Internet portal like Yahoo! provides marketers the ability to establish a complete view of key consumer groups, including the integration of offline and online activity. It allows the marketer to define the consumer behavioral profile from the consumer panel data and then identify which search words and pages within Yahoo! these consumers are visiting.

A Case Study

ACNielsen, in conjunction with Yahoo!, has begun implementing a program to evaluate the effectiveness of online promotional campaigns by analyzing the impact of this online activity on actual offline sales among key consumer segments. The service, called Consumer Direct, allows marketers to identify, communicate and evaluate campaigns of this nature.

The goal of Consumer Direct is to partner with marketers to help plan and personalize online campaigns. With the ACNielsen Homescan consumer panel, marketers can understand offline purchase data, brand loyalty measures, category usage information, and promotional sensitivity data. Yahoo! has over 100 million users, which provides online behavioral information (both declared and observed from search, content and ad activity), online transaction data, user demographics (age, gender, geography) and scalable platforms, including ads, promotions and surveys.

With a recent Consumer Direct program for a major packaged goods manufacturer, using a permission-based subset of the Homescan consumer panel and creating a “link” between these panelists and Yahoo! visitors, we were able to identify-communicate-evaluate the effectiveness of an online campaign on offline purchase behavior [See chart 3].

The “linked” panel allows the segmentation of consumers who currently purchase a category or brand to their online activity (i.e., sites visited). These segments were then evaluated based on the client’s marketing goals and objectives. The client worked with ACNielsen to define high potential consumers—in this case, Heavy Category Users.

Next, ACNielsen used Yahoo! information to determine the online profile that best distinguished those in this “desired” consumer segment versus those who were not part of this group. The distinguishing characteristics of these Heavy Category Users translate into a set of rules that statisticians call a “scoring algorithm.” The scoring algorithm was then applied to the entire Yahoo! visitor file, and the result was a ranked list of all Yahoo! visitors, from those who behaved most similarly to these Heavy Category Users.

This ranking allowed for the capability to communicate with this select group of Yahoo! visitors at the top of the list, to receive a specific online promotional message through Yahoo! (a random group was held out as a control group and did not see any of the client’s branded media on Yahoo!). The usual online metrics (clicks, page views, promotional site interaction, etc.) were also tracked [See chart 4].

In this case study campaign, a special promotional web site was developed by the client to provide information to category users. The objective of the campaign was to drive the Heavy Category Users to this special promotional site and increase brand purchases. Target consumers received 90 million impressions and 13 million unique visitors visited the site.

The Results

Actual offline purchase data was then analyzed to determine the effect the campaign exposure had on actual sales, by comparing those consumers “exposed” to the program versus those that were “not exposed.” In this case, the Exposed group had a sales lift of 1.28 while the Non-Exposed group had a sales lift of 1.16, a statistically significant difference of 9.6%.

In this case study, the Yahoo!-delivered campaign had a positive, statistically significant impact on offline purchases. Dollar purchasing increased 76% and penetration increased three times among the Exposed group. Consumer Direct increased targeting efficiency by 73%. Yahoo! online targeting resulted in category reach of 24%, while the Consumer Direct targeting resulted in category reach of 41%. Dollar sales attributed to the campaign were $1.8 million, and the contribution of $620,000 resulted in a return on investment of 40%. And finally, 100,000 new brand buyers were added to the franchise.

The Internet is now a medium that packaged goods marketers must incorporate into their total marketing mix. Although there has been much criticism of web site development costs and online media spending, if a creative and analytical approach is taken and there is an integration of online campaigns with other offline media/promotional programs, significant sales and brand development successes can be delivered through direct-to-consumer marketing programs on the Internet. Most importantly, return-on-investment measures of media spending can now for the first time be measured for these online campaigns.

By; Robert L. TomeiSVP, Chief Marketing OfficerVNU Marketing Information Group