You are here: Research Studies -> gemiusEffect -> Research Methodology
gemiusDirectEffect - measuring the delivery, instant response, delayed response post-view, actions and conversions

Methodology of gemiusEffect

Analyzing the effectiveness of an advertising campaign must take place in a neutral, objective environment. Each of the three modules that are used for gemiusEffect is independent, objective and neutral. Although integrated through the gemiusTraffic platform, all three modules are conducted using different methodologies to address their different issues.

  • gemiusDirectEffect

gemiusDirectEffect is executed through site-centric measurement using Gemius' proprietary gemiusTraffic platform. gemiusDirectEffect is research on the direct response to your online advertising campaign and directly measures Internet user behavior.

The survey is executed by connecting gemiusTraffic scripts to emitted creative and on the advertiser's web site. The scripts gather data on each emission and clickthrough by Internet users. Thus, we gain data on how many times an Internet user clicked on the campaign and how they behaved after clicking (or if they did not click how they behaved).

  • gemiusBrandingEffect

The research tool used for gemiusBrandingEffect is a web-survey displayed to a random sample of respondents online. Sample selection is done using an experimental methodology on two sub-samples: a Control/Exposed methodology.

The first step in the research is to connect specially-prepared gemiusTraffic scripts to the online campaign's creative and on the web site being advertised. These scripts then register each emission of the advertising creative and every clickthrough to the advertiser's web site.

The gemiusTraffic scripts work to recruit a random sample of Internet users who have been exposed to the campaign (the Exposed Group) and a sample that has not been exposed (the Control Group). A comparison of survey results between these two groups makes it possible to analyze the impact (lift) on key indicators that the online campaign caused. Identification of users who were exposed to the campaign and those who were not is provided for thanks to the gemiusTraffic scripts and Gemius' site-centric gemiusTraffic platform. Surveys are emitted to a random sample of both groups, recruiting every nth user who visited the given web site.

  • gemiusProfileEffect

The gemiusProfileEffect study is based on the gemiusAudience study conducted by Gemius in order to measure the socio-demographic profile of the users visiting over 350 web sites. Thanks to the gemiusAudience study, Gemius has developed a quasi-panel of Internet users identified through the use of cookies.

The gemiusAudience study implements a unique analytical algorithm to maximize the value of data: an Internet user who fills out a survey on site A, when visiting site B, brings his socio-demographic profile with him to site B. Thanks to this, a user who fills out a survey on one site builds the total sample of respondents across all sites that he visits.

The gemiusProfileEffect study uses the dataset gained from the gemiusAudience study. The sample size of the study is directly related to the campaign's reach (because the campaign's reach affects the likelihood that the campaign reached a user who has filled out a gemiusAudience survey). Gathered data are behaviorally weighted to the population that was exposed to the campaign, adjusting for sample selection bias. Thanks to this process, we are able to analyze data at the level of overall campaign audience as well as down to narrow groups such as those exposed to specific creative or even creative/placement combinations.

If you would like additional information on gemiusEffect's methodology, feel free to contact Gemius