The European Society for Opinion and Marketing Research (ESOMAR) is a non-profit organization founded in 1947 to promote the value of the market, opinion, and social research and data analysis. It provides honest and professional advice, for example, through its well-known international code of conduct, written in conjunction with the International Chamber of Commerce (ICC).
ESOMAR's 28 Questions' primary objective is to increase transparency and awareness of the critical issues in online market research and help companies decide whether a digital market research company is in their best interests. In detail, ESOMAR's 28 questions deal with topics such as the company profile, sample sources, or policies & compliance.
1. What experience does your company have in providing online samples for market research?
BitBurst was founded in 2015 to develop mobile apps. Since 2018, we have been focusing on the digital market research industry, reaching over 2.5 million people worldwide with our platforms (apps, websites, etc.) and partner platforms, completing over 1 million surveys in 2020, and experiencing solid revenue growth since inception.
2. Please describe and explain the type(s) of online sample sources from which you obtain survey participants (e.g., databases, actively managed research panels, direct marketing lists, social networks, or web intercept samples).
Our survey participants' primary source is our apps/websites that allow people to participate in rewarded surveys. Besides, we are continuously building a network with selected partners who can integrate our surveys into their apps using our BitLabs platform. Our partners operate in the automotive, cryptocurrency, gaming, entertainment, or lifestyle sectors, among others.
3. If you provide samples from more than one source: How are the different sample sources combined to ensure their validity? How can this be replicated over time to ensure reliability? How do you handle the possibility of duplicating participants across sources?
Our channels, and therefore samples, are merged to a common point where we use UID (unique identifier) to verify that a user enters each survey only once. Also, we work with our partner Sample-Chain, which systematically prevents fraud attempts with multiple participations.
4. Are your sample sources used exclusively for market research? If not, what else are they used for?
Our platforms are used exclusively for market research purposes. Some formats also conduct market research solely for our partner platforms and those where market research is a sub-business. For example, other operators integrate our surveys into their apps, whereby their users can earn advantages in a game, free minutes for their cell phone contract, or temporary premium memberships by participating.
5. How do you reach survey participants who are difficult to reach via the internet?
With over 2.5 million users generated through our platforms, we can efficiently survey hard-to-reach participants. To motivate this target group to participate in a survey, we can also increase their reward individually.
For special projects (see question 6), we can access marketplaces in addition to our usual channels and thus fulfill even very complicated quotas.
6. If you have to supplement your samples with those of other providers for special projects: How do you select these partners? Do you inform customers in advance if you use third-party providers?
We only use samples from our platforms and those of our selected partners. As described in question 5, a customer may want hard-to-reach participants or unusual quotas. In consultation with our customer, we have the option, in this case, to fall back on selected marketplaces with which we could again increase our reach.
7. What steps do you take to obtain a representative sample of your target group?
To ensure that our customers get as representative a picture as possible of their intended target group, we advise on important parameters that the respective sample should cover. The statistical offices of the respective countries serve as a basis, providing us with information on distributions according to, for example, gender, age, or region. Based on the desired parameters, we can offer precisely fitting numbers of participants for the survey. To further support our surveys' representativeness, we also calculate minimum participant numbers based on generally accepted statistical formulas.
8. Do you use a survey router?
Yes, with our survey routing, we offer an effective option that is very different from the usual practice of offline market research: Our users are already profiled by, e.g., age, gender, income, etc. before participating in the survey and are only routed into a study if their profile matches the requirements.
9. If you use a survey router: Please describe the assignment process. How do you decide which surveys are considered for which participants? Based on what priorities are participants assigned to surveys?
Our users qualify before survey participation by answering questions about their profile (age, gender, income, etc.). If a user is eligible for a survey, he/she is allowed to participate. Surveys are arranged and played out by the best user experience to give our users a better experience with our platforms.
To increase survey quality and user interest, we only display surveys that suit the participant.
10. If you use a router: What measures do you take to compensate for or prevent bias due to the use of a router? How do you measure bias and how do you report it?
Bias is prevented at our company by a significantly high survey and user volume, the extensive mix of target groups, or even the purely random assignment of surveys to users. One advantage of our systems is that no specific categories or topics are routed, and users thus receive questions randomly based on their general profile characteristics. This system prevents users from, for example, not receiving a survey from a similar topic in which they have already completed a survey.
11. If you use a router: Who in your organization sets the parameters of the router? Is it a dedicated team or individual project managers?
Our Integrations department sets the router parameters, a team of developers who are the only administrators who can change these settings. The routing is based on product-specific assignment algorithms to improve the user experience.
12. What profile information is collected about panel members? For how many users is such data collected and how often is it updated?
On the one hand, we collect general information such as gender, age, place of residence, or income. On the other hand, we can also ask for survey-specific qualifications such as religious denomination, product interests, or others.
Each user provides this information voluntarily and thus qualifies for more surveys. We also collect high-level data that does not need to be updated (e.g., year of birth instead of age); on the other hand, we ask our users at regular intervals whether their data is up-to-date. For security reasons, users must notify us of their writing changes and cannot change them themselves. This step is to prevent participants from "segment hopping" and falsely participating in surveys by changing their profile data.
13. Please describe the invitation process for your surveys. What offer do participants get to participate in a survey? What information do they receive about the project as part of this process? What other methods of invitation (besides direct solicitations) are used to target participants to specific surveys (or to a router)? Note that not all invitations are done by email.
On our platforms, a user gets the surveys that are appropriate for them on their home page. If a user finds a survey attractive, he proactively selects it, participates in the survey, and receives a reward. For partner platforms with market research as their primary purpose, the invitation process is similar. For platforms that do not conduct market research as their primary purpose, the invitation is made, for example, simply via a pop-up window that informs the user about participation in a survey.
If a user decides to take part in a survey, we ask him or her as much as possible about the purpose of the survey and the relevance of his or her individual opinion before the survey begins.
We also send notifications in our apps, for example, to inform our users about recent surveys or other topics.
14. Please describe the incentives offered to respondents for participating in your surveys. How does this vary by sample source, interview length, and respondent characteristics?
Through our platforms, participants receive reward points to complete a survey, which they can either cash out or exchange for vouchers (Amazon, Google Play Store, Apple, etc.). On our partner platforms, users receive rewards, e.g., in the form of unlocks or app-specific reward points.
To reach users with complicated characteristics or to motivate participants for long surveys, they receive higher rewards.
15. What information about a project do you need to give an accurate assessment of feasibility using your resources?
Due to a very high outreach to many different people, we can carry out almost any project. To estimate the effort required for a project, we need
i. Demographic data (country of origin, age, gender, etc.),
ii. Special qualifications (such as product interests), if applicable,
iii. The number of participants and quotas to be met by the target group,
iv. Length of survey or number of questions, and
v. The question format.
16. Do you measure respondent satisfaction? Is this information made available to customers?
Yes, we measure respondent satisfaction via the respective completion rate or dropout rate. We integrate feedback questions at the end of self-launched surveys. Users can report inappropriate surveys, and our platforms are rated via the Google Play Store or App Store, for example.
17. What information do you provide to your customers after project completion?
After the survey is completed, we perform a data check, prepare the data set, and then send our clients the desired format results (e.g., graph, Excel, CSV, or SPSS). Besides, our customers have the option of receiving evaluations filtered specifically according to target groups (e.g., what did a certain age or income group answer?).
18. Who is responsible for data quality checking? If you are, do you have procedures in place to reduce or eliminate undesirable behaviors within the survey, such as (a) random responses, (b) illogical or inconsistent responses, (c) increased selection of non-responses (e.g., "Don't Know "), or (d) speeding (completing the survey too quickly)? Please describe these procedures.
A key feature of our service is that we perform data quality checks ourselves. We try to avoid arbitrary, illogical, or inconsistent answers, for example, by using logical strings and double but differently worded questions. To reduce the selection of non-answers, we offer the participants as many answer alternatives as possible and give their answers with free fields. To prevent speeding, we first test our survey ourselves before publishing it and set a minimum time that a participant needs to complete the survey. Responses that fall outside this tolerance range are removed from the dataset. In addition, we also work with our partner Sample-Chain on this, prevent link manipulation with an SHA-1 encryption process to verify response sources, and solicit honesty from participants at the beginning of a survey.
19. How many times can the same person be contacted to participate in a survey within a given time period, regardless of whether they respond to the contact or not? What limits are observed? How does this vary between your sample sources?
Our users are generally not contacted directly by us but proactively self-select surveys through our platforms. However, we can control the number of surveys offered to a single participant. For more details on how often a user can participate in surveys, see the answer to question 20.
20. How many times can the same person participate in a survey within a given time period? How does this vary between your sample sources? How do you manage this within categories and/or time periods?
To avoid frequent user participation, we classify those with regular survey participation and an above-average conversion rate as suspicious and exclude them from our systems after checking if necessary. Other points that contribute to ensuring that a user receives surveys from a wide variety of categories or does not move too frequently in the same topics are the very high survey and user numbers, the completely random assignment of surveys according to general profile characteristics, or the very broad mix of our panels.
In consultation with our customers, we can also filter users by the frequency of their survey participation in a given period and exclude them from a survey if necessary.
21. Do you also collect individual data, such as participation history, date of joining, source, etc. for survey participants? Can you provide your client with an analysis of such individual data?
Yes, we collect data such as entry date, completion time, panel source, etc., and can provide our clients with an analysis of this. Through this data, we can accurately analyze the behavior of our users and incorporate it into the quality of our systems and concrete surveys.
22. Do you have a procedure to verify the identity of respondents? Do you have procedures in place to detect fraudulent participants? Please describe how procedures are implemented at the time of sample source registration and/or entry into a survey or router. If you offer B2B samples, how are procedures implemented there, if applicable?
Among other things, we use two procedures to identify users uniquely: on the one hand, they have to authenticate themselves with a social media account (Facebook, Google, Apple, Twitter), and on the other hand, our users need a phone number that can only be used once. The latter is handled by one of our partners who specializes in phone number verification.
23. Please describe the "opt-in to market research" processes for all your sample sources.
First, before participating in a survey, each participant must confirm our T&Cs. Additionally, survey selection is proactive, so the user expresses consent not by responding to an invitation but by proactively selecting the survey. As a third instance, a participant will be educated about the particular survey before being surveyed and will be required to consent to that as well.
25. Please describe the measures you take to ensure privacy and data security.
At BitBurst, we handle our users' data responsibly and in compliance with important legislation, such as the General Data Protection Regulation (GDPR) or the guidelines of the European Society for Opinion and Marketing Research (ESOMAR). We always try to give our users full transparency and control over their data, do not sell or market personalized data to third parties, and only give employees the access to data of our customers and users necessary for their work.
26. What practices do you follow to determine whether commercially sensitive customer data or materials are presented in online market research?
First, we determine how sensitive the customer data is and which survey format makes the most sense for the specific purpose. If the client requests it, we advise our participants at the outset of the data's confidentiality and possible legal consequences. If a participant does not agree to the confidentiality agreement, they will not participate in the particular survey.
27. Are you certified according to a certain quality system? If yes, which one(s)?
BitBurst is among the top performers in the rankings of established panel marketplaces. For example, we achieved second place among >100 providers in the 2020 Lucid Quality Rankings.
28. Do you conduct online surveys with children and young people? If so, do you adhere to the standards provided by ESOMAR? What other rules or standards (e.g., COPPA in the U.S.) do you adhere to?
We conduct surveys exclusively with adults and "young people" aged 16 and older, according to ESOMAR guidelines (see: Guideline on Research and Data Analytics with Children, Young People, and Other Vulnerable Individuals). We take care to protect the younger participants by not publishing surveys with inappropriate content for their age (e.g., pornography or the consumption of alcohol).