Our Quality Management
GMO Research & AI manages quality control in three ways: panel quality, data quality, and information security quality to maintain our high standards and the health of our online research platform.
1. Panel Quality
Panel quality management is critical to ensure the reliability of survey results. The inclusion of respondents of poor quality can bias survey results and make it impossible to attain accuracy. GMO Research & AI believes that it is our responsibility to manage, maintain and ensure the quality of respondents.
Quality Management of the Proprietary Panel
Registration
Our own panels require e-mail address, SMS and phone number verification for membership registration.
Information Update
We update members’ attribute information once a year through regular attribute acquisition surveys. Through this regular data collection, we acquire accurate survey results.
Member Acquisition Optimization
We avoid causing response bias in our surveys by obtaining a wide range of members in terms of various attributes (gender, age, and place of residence).
Quality Management of the Cloud Panel
No Duplication
We pre-investigate and manage duplication of panelists using cookies and e-mail addresses to avoid duplicated respondents and response bias.
Inactive Panelists Measures
To conduct accurate and effective surveys, we take measures to stop sending surveys to panelists who have not responded for a certain period of time.
Anti-Fraud Measures
We check our survey response data for detection of inconsistencies of basic attributes: straight-lining, bot respondents, speeders with extremely short response time, duplicated cookies, etc. We do not send any more surveys to the respondents who meet certain standards for fraud. We also conduct quality check surveys periodically to control the respondents who may have answered untruthfully or fraudulently.
Educational Survey and Trap Survey
We periodically conduct educational surveys with panels to encourage them to respond appropriately and improve their awareness and knowledge of surveys. In addition, we detect inconsistencies of basic attributes and contradictory responses by conducting trap surveys that include trap questions. We do not send any more surveys to the respondents who meet certain standards for fraud.
Automatically detects and eliminates 'Ghost Completes'
GMO Research & AI's proprietary technology analyses fraudulent responses and has developed and implemented a function to automatically detect and eliminate 'Ghost Completes'.
* Ghost Completes are fraudulent internet research surveys in which respondents are hacked to disguise the fact that they have completed a survey, even though they have not, in fact, done so, in order to gain points or other rewards.
2. Data Quality
GMO Research & AI provides unbiased and representative data of our survey results through the data quality management process. We maintain a high standard of quality by establishing technology-based quality control and rigorous data checking, operated by our professional research operators.
Before Surveys
- Previewing and checking the questionnaires
- Managing survey logic to avoid contradictory responses
- Double-checking the questionnaires with a third party
- Homogenizing the quality of service by unifying the workflow of each operator
During Surveys
- Eliminating duplication of panelists using cookies
- Calculating the appropriate number of survey distributions automatically with our technology
- Managing survey logic to avoid fraudulent respondents
- Confirming response status and collecting survey data in real-time
After Surveys
- Raw data cleaning
- Eliminating inconsistent responses, speeders’ responses, and open-ended responses that are obviously fraudulent, etc.
- Systematically detecting and excluding fraudulent data through simple aggregation
In addition to the above, GMO Research & AI is engaged in a wide range of activities and initiatives for maintaining data quality: data cleaning according to the characteristics of the panels and the purpose of the surveys, designing and programming in consideration of the multi-device era, controlling the volume of questions, industry awareness activities, and many others.
We have also implemented a comprehensive data management system for the information provided to clients. As the world's 9th largest consumer panel provider, we continue to improve quality and technical infrastructure.
* The above is a partial excerpt of the process we are following for our full-service survey.
* The process varies depending on the service provided.
3. Information Security Quality
All consumer panels adhere to the quality control standards of JMRA (Japan Marketing Research Association) and ESOMAR (the European Society for Opinion and Marketing Research).
We have obtained and follow the P-Mark, a Japanese certification for personal information protection, and the ISMS (ISO/IEC 27001: 2013), a global certification for data security. We have also appointed a team of data protection officers to ensure that we comply with GDPR, CCPA, and other area-specific laws and regulations.
In order to comply with the China PIPL (Personal Information Protection Law), the new data privacy law which came into effect on November 1, 2021 in China, we established a task force and have made every effort to respond accordingly. We constantly monitor the trends in personal information protection laws and other legal developments around the world and promptly respond to amendments or enforcement of such laws.
ESOMAR 37
ESOMAR has developed 37 questions to "provide a standard set of questions a buyer can ask to determine whether a sample provider’s practices and samples fit with their research objectives."