Determining Project Feasibility, Mode, and Pricing
What services does GeoPoll offer?
GeoPoll is a full-service research firm, offering end-to-end support for a variety of research projects. We can assist with everything from questionnaire design to data analysis, and often create custom research solutions for our client’s needs. Common services we offer include:
- Indicator measurement and tracking: GeoPoll is able to track humanitarian indicators including food insecurity, nutrition, stability, governance, and other indicators in real time among remote populations. For consumer groups, we also track brand health, net promoter score, and other key metrics.
- Public perception and knowledge: GeoPoll provides insights into public perceptions and knowledge that can enable partners to design tailored activities, support adaptive program management, and measure effectiveness of programs and campaigns.
- Campaign evaluation: GeoPoll can evaluate the effectiveness of advertising or other campaigns by conducting research before, during, and after planned activities.
- Panel studies: GeoPoll can create and maintain ongoing panels using the GeoPoll database and has successfully run daily, weekly, and monthly panel studies on a variety of topics.
How does GeoPoll determine the feasibility of my project?
GeoPoll has an entire support division to assist in reviewing the feasibility of a project and guide you as you decide on survey mode. Our feasibility team reviews the project requirements and compares them to our history of performance and can even run no-cost tests to confirm a project is feasible, saving you time and money to know if your survey project can go forward.
For simple one-off ad-hoc surveys, to determine the feasibility of a project we need to know the country(s), date, time, and frequency, the preferred survey mode, the sample size, the sample source (GeoPoll or client), the sample demographics and any incidence, the language, and the number of questions. For more complex panel projects, we may also need to take into account the panel methodology, sample churn, recent response rates, and the duration of project.
What survey modes does GeoPoll use and what is the difference between each?
GeoPoll works across multiple modes, including SMS, live calling, in-person, and internet-based surveys. Our research experts will work with you to determine the most appropriate mode to contact your target population.
- Short Message Service (SMS, or text message): SMS surveys are self-administered surveys which are sent to respondents one question at a time (see an example here). SMS surveys does not require respondents to have internet connectivity or airtime credit, and the survey invitation is received even if phone is turned off or out of service range when sent. Respondents can also answer the questions at their own leisure, ensuring privacy when taking surveys with sensitive questions.
- Computer Assisted Telephone Interviewing (CATI, or live calling): CATI surveys are administered through experienced call center interviewers based in the country they are calling. GeoPoll interviewers go through rigorous trainings for each project they administer and are overseen by dedicated in-market call center supervisors and GeoPoll’s senior team. GeoPoll’s proprietary mobile application for CATI interviewers facilitates interviewer flow, data collection, and interviewer oversight and can be utilized on any smartphone or tablet. Features of our CATI Application include offline mode, call recording, multi-lingual support, and more.
- Computer Assisted Personal Interviewing (CAPI): Built on the same system as GeoPoll’s CATI application, GeoPoll’s CAPI application facilitates face-to-face data collection in the field, removing the need for paper questionnaires and manual data collation. The CAPI application does not require interviewers to have internet access when conducting surveys and provides interviewer metrics for overseeing projects.
- Interactive Voice Response (IVR): IVR is a self-administered survey method in which pre-recorded voice surveys are initiated via phone call, and respondents key in answers on their dialpad. IVR is useful for reaching illiterate populations through voice calls without the need for training call center interviewers, and can therefore send a mass amount of surveys at once through automated software.
- Mobile Web: Mobile Web surveys are self-administered surveys which require that respondents have internet-capable phones with an active WiFi or data connection. Respondents are sent an initial SMS message with a link to a survey, which drives respondents to their web browser to complete the survey. Mobile web surveys support GPS location, video/picture questions, and matrix/ table/ grid format questions.
- Mobile Application: Mobile app surveys require a smartphone, although the mobile application does allow for offline use. Respondents who download the GeoPoll mobile application are sent notifications to take part in surveys through GeoPoll’s mobile app. The mobile app includes, with respondent consent, additional capabilities such as GPS location, picture taking and passive data collection.
- Focus Groups: GeoPoll can implement focus group research through mobile chat groups overseen by an experienced moderator. Focus groups allow for sharing of video and picture content, structured discussions, and other tasks and can be maintained over a longer period of time than in-person groups.
What does “multi-modal” mean?
GeoPoll’s technology platform supports all of the survey modes described above, meaning a project can engage multiple modes to reduce costs, increase speed, and improve the representation of the groups trying to be reached. For example, projects can incorporate both face-to-face data collection via CAPI and follow-up via CATI voice calls or SMS surveys. Data for all modes is fed into GeoPoll’s central system where it undergoes quality checks before being packaged in the desired data delivery format.
Do all GeoPoll surveys require respondents have access to a mobile phone and how do you reach those without mobile phones?
No. GeoPoll leverages mobile technology to collect survey data, and often this means reaching respondents via their own mobile phones. As mobile penetration grows around the globe, this allows us to collect data quickly and remotely at a wide scale. Additionally, GeoPoll uses several techniques to reach targeted populations and nationally representative samples via mobile modes. However, we do implement face-to-face surveys via CAPI which allows us to reach those who do not have access to a mobile phone, such as those in the lowest income groups.
Do respondents need a smartphone to participate in GeoPoll surveys?
The majority of GeoPoll surveys do not require a smartphone or internet access. GeoPoll’s Mobile Web surveys are administered via a link and require a basic mobile browser and internet, but do not require a smartphone. The only mode that requires a smartphone is GeoPoll’s Mobile Application, which has been built to work on a wide range of devices and operating systems, including those which are several years old.
What are "trigger" surveys? How do they work?
Trigger or text-in surveys describe a survey in which a respondent sees an SMS shortcode and keyword, for example on a product package, in an advertisement, or in another physical location, and texts the keyword to the shortcode to initiate a survey. This method is useful for gathering feedback from specific groups, such as those who buy a certain product or visit a store. Once the respondent has triggered in to a survey, it proceeds in the same way a regular SMS survey would.
What is GeoPoll’s pricing structure?
GeoPoll’s pricing is determined by multiple factors including survey mode, respondent targeting and sample size, questionnaire length, and data delivery format. To request a quote for your project, please contact us.
GeoPoll Respondent Database
What sample frames does GeoPoll use?
The creation of a sampling frame for GeoPoll projects depends on client needs, project specifications, and other factors including survey mode. While sampling frames are unique for each project, there are a few common sampling frames that we use which are outlined below.
- Mobile subscribers within a certain country: Sample frames for our studies are often those who have access to a mobile device within each country. GeoPoll reaches mobile subscribers in two primary ways: Partnerships with mobile network operators which enable us to call or send messages to their opted-in subscribers who make up the GeoPoll database, and Random Digit Dialing (RDD). We can also utilize online or offline advertising to gather additional respondent phone numbers when necessary.
- Census data: GeoPoll also relies on census data and census estimates both to inform nationally representative demographic breakdowns and to create sample frames when conducting in-person research. The availability of up-to-date census data varies by country and requires a researcher to understand what information from reputable sources is available. One resource that can be used to look at each country’s local bureau of statistics and at the U.S. Census Bureau’s International Data Base.
- Aid Beneficiaries, Customers, or Employees: When working with international development clients, GeoPoll is able to survey aid beneficiaries if given their contact information. This requires organizations to provide GeoPoll with a list of beneficiaries’ phone numbers or other contact information. GeoPoll can also survey customers or employees of consumer brands or other organizations if given contact information, or can survey visitors to retail locations via a trigger-in technique.
What sampling techniques can GeoPoll use?
GeoPoll can use a variety of sampling techniques but most often utilizes probability sampling methods including simple random sampling and stratified random sampling. If no demographic quotas exist, GeoPoll can use simple random sampling within the specified country. If a project does have quotas for demographics, we will utilize a stratified random sampling approach in which we randomly target GeoPoll database users who are pre-stratified by demographic characteristic such as gender, age, and location in order to get a sample that is representative of the general population or the agreed upon sample universe.
What is Random Digit Dialing (RDD) and how does GeoPoll’s RDD process work?
Random Digit Dialing (RDD) is a type of probability sampling in which phone numbers are randomly generated using a software system and used to create the sample for a project. By using an RDD approach, the sample frame becomes all mobile subscribers in a country, rather than mobile subscribers from specific mobile network operators or from another smaller list. GeoPoll uses an intelligent RDD process in which we identify the phone number prefixes used by each mobile network operator in a country and generate numbers based on those formats. We then use various techniques to verify RDD numbers, removing invalid numbers from our lists. Finally, we conduct tests on the verified RDD lists to determine how well the sample is performing and can generate additional sample if needed. To learn more, read this post.
What is the GeoPoll respondent database? How does GeoPoll get access to respondents’ phone numbers?
The GeoPoll database is a database of over 250 million mobile phone subscribers, built through our partnerships with over 85 mobile network operators around the globe, in addition to other recruitment methods such as online advertising or face-to-face recruitment. The GeoPoll respondent database is further segmented into two primary groups, active database members, and inactive database members:
- Active Database Members are respondents who have taken at least one survey with GeoPoll and opted-in to be contacted for further surveys. GeoPoll has stored demographic information including age, gender, and location for our over 5 million active database members, and most of our projects sample from this group as it allows for pre-stratification of our database.
- Inactive Database Members have not taken a GeoPoll survey but are available to use in certain circumstances – for example if a survey has very specific targeting and we are unable to reach enough sample through our active database.
How does GeoPoll verify respondents’ demographic information?
For self-administered modes including SMS and web link, GeoPoll takes several steps to ensure demographic information is correct. When a respondent first opts-in to the active GeoPoll database, they take an onboarding survey which collects demographic information. This data is stored in the GeoPoll database, and in each subsequent survey a respondent takes they are re-asked key demographic questions. GeoPoll cross-checks answers for each survey with the originally provided answers to ensure demographics do not change over time, in which case a respondent may be removed from the GeoPoll active database. In addition to regular re-checks of demographic data, GeoPoll can reach those who have conducted surveys via SMS or web link by voice call to further verify self-reported information. These measures, in addition to other quality controls we have in place, ensure high quality, accurate respondent data.
For interviewer-administered surveys, interviewers ask respondents to report their demographic information based on the demographic questions included in each questionnaire.
How does GeoPoll determine respondent location/urban vs rural locations?
Respondent location for remote methodologies is self-reported but can be verified in several ways. First, GeoPoll regularly re-verifies demographic information including location to ensure the respondent’s location is up to date. For certain modes, such as the GeoPoll Application, we can use GPS to further verify a respondent location. GeoPoll’s standard location targeting is done at the first administrative district of a country (ADM1), but we can ask additional questions to gather information on ADM2 and ADM3. GeoPoll also gathers location data through a basic question asking respondents if they are in an urban or rural environment.
Can GeoPoll reach nationally representative samples?
Yes, GeoPoll reaches nationally representative samples through several techniques. For mobile-based methods including voice call, SMS, and web link, GeoPoll generally uses population estimates by age, gender, the first administrative political geographic boundaries in order to create a sample framework that matches the overall population. However, GeoPoll is happy to adapt frames with other demographic variables during the planning stages of the project. We then use a stratified random sampling technique to ensure we gather enough respondents from each demographic group to match the census estimates. Depending on the project need, we can also include quotas for education, income or living standard measure, and additional demographics. This technique enables us to create a nationally representative sample even when surveying only those with mobile phones.
While access to mobile phones has become more widespread across all parts of the globe, GeoPoll recognizes that the mobile population does not always perfectly match the national population. For projects that require we reach the lowest income groups or those without mobile phones, we can also deploy face-to-face methods to reach part or all of the desired sample.
Can GeoPoll utilize client-provided sample instead of the GeoPoll database?
Yes, GeoPoll can use client-provided sample instead of or in addition to the GeoPoll database or other methods like RDD. Client provided mobile phone numbers work well when the recipients on the other end are expecting to receive and participate in the studies. When surveys are sent to “cold” numbers who are not provided advance notice by the partner organization, response rates are much lower than those seen with active GeoPoll database members so clients should be prepared to provide large sample files.
How does GeoPoll use pre- and post-stratification (quotas and weighting) to achieve the desired sample composition?
Active members of the GeoPoll database are stratified by demographic characteristics including age, gender, and location, and can be targeted based on these attributes. GeoPoll uses pre-stratification techniques as part of our stratified random sampling approach to create samples that are nationally representative or target specific demographic groups.
If our sampling approach isn’t able to achieve the desired sample breakdown due to data collection challenges, budget issues, or other factors, GeoPoll is also able to use post-stratification or weighting in order to bring the data more in line with the desired sample characteristics. GeoPoll uses cell-based weighting and raking techniques most often, but our research team can work closely with clients while performing weighting in order to achieve an optimal dataset.
Can GeoPoll recruit and maintain ongoing panels?
Yes, GeoPoll has experience recruiting and maintaining panels for ongoing data collection. GeoPoll Audience Measurement has a panel in each market it is run in, and we can create custom panels of specific groups, such as farmers or certain demographics, based on client need.
Respondent Consent and Privacy
What is the opt-in/consent process for the GeoPoll database and individual surveys?
Are respondents in the GeoPoll database used for anything other than market research?
No. GeoPoll abides by ESOMAR guidelines regarding market research respondents and we do not engage in direct marketing or any other non-research activities with our database members.
Does GeoPoll conduct research with children under 15?
In line with market research best practices GeoPoll does not conduct surveys with those under the age of 15 unless mobile numbers provided also confirm consent from a parent or guardian.
What steps does GeoPoll take to ensure data protection and security?
GeoPoll takes data security extremely seriously, and takes several steps to ensure data is protected including utilizing encrypted, high-availability servers stored within a secure datacenter, and implementing strict access control for all employees and contractors. All systems and servers are securely backed up on a nightly basis to encrypted off-site storage with an RPO of 24 hours, and servers and employee laptops use hard drive encryption and run antivirus / anti-malware software. Additionally, all employees and contractors undergo background checks prior to employment, and all systems and services undergo penetration tests on a quarterly basis.
To request GeoPoll’s full Data Protection Policy, please contact us at [email protected]
How does GeoPoll protect respondents’ personal identifying information?
How does GeoPoll comply with IRB and other ethical research standards?
GeoPoll takes respondent privacy and research ethics seriously and follows industry guidelines surrounding research ethics, including those set by the European Society of Opinion and Marketing Research, of which we are an active member. We also comply with Institutional Review Board research protocols and have experience working with clients to gain IRB approvals when projects require them.
Survey Design and Implementation
Does GeoPoll assist clients in research methodology and questionnaire design?
Yes, GeoPoll’s team can assist in designing a research project from start to finish, including helping you decide on your survey mode, methodology, and sample frame, incorporating mobile research best practices into your existing methodology, and designing questionnaires. Our dedicated questionnaires team can design surveys from scratch, assist with transitioning a face-to-face questionnaire to be used in a mobile mode, or simply perform a quality check of your own questionnaire before it is scripted in GeoPoll’s system.
What languages does GeoPoll support and do you assist with questionnaire translation?
GeoPoll has administered surveys in over 40 languages and our platform supports most languages and scripts. We also support multi-lingual surveys which give respondents the option of which language to proceed in. GeoPoll can provide questionnaire translations which are checked for accuracy by our local partners.
Does GeoPoll have its own scripting platform? Can I use another scripting software?
Yes, GeoPoll has its own scripting platform which is the backbone of all surveys conducted via SMS, voice call (CATI or IVR), and face-to-face (CAPI). GeoPoll’s scripting platform is built to support many question types and integrates seamlessly with the rest of the GeoPoll’s systems. After a questionnaire is finalized, GeoPoll’s team scripts it within our platforms and will then send a test survey to the client for final approval before it goes into the field.
For link-based surveys, if you have a pre-scripted link using another software we are happy to utilize that link to send surveys to GeoPoll’s database via mobile web or mobile application. In these cases, GeoPoll is providing sample only, as we do not have control over survey scripting or data analysis.
Does GeoPoll support advanced question routing and skip patterns?
Yes, across all modes the GeoPoll platform has developed the capabilities to run nearly every kind of question that a common paper or online survey can run, including single choice, open-ended, select-all-that-apply, matrix questions, and more. Our platform also supports complex logic, including routing, looping, filtering, dynamic question types, and randomization of answer options and question order.
What are some key considerations when transitioning face to face data collection to live calls (CATI)?
Live calls conducted through Computer Assisted Telephone Interviewing (CATI) are a popular alternative to face-to-face research especially in remote locations or when in-person research is infeasible due to disease outbreak, violence, or other factors. Key considerations for transitioning a face-to-face project to CATI include keeping the interview to 20-25 minutes long and using certain question types such as open-ended questions sparingly.
If the CATI project will use client-provided phone numbers, we recommend double checking the numbers, sharing the mobile network operator along with the phone numbers where possible, and alerting participants beforehand that they will be receiving a voice call. GeoPoll’s team is experienced in transitioning face-to-face projects to CATI and will advise you on other considerations throughout your project.
How does GeoPoll train and monitor interviewers for interviewer-administered modes?
Interviewers for GeoPoll’s CATI and CAPI surveys typically have multiple years’ experience with survey enumeration, and go through a 3+ day extensive, project-specific training before starting data collection on any new project. Within this training, GeoPoll managers review the project questionnaire, key answer definitions, guidance for probing on specific questions, and more. Interviewers then go through a practice exercise with the full questionnaire and are provided feedback by managers. Finally, during survey production GeoPoll monitors interviewer progress through our quality assurance dashboard, which enables us to see any data anomalies and quickly address them.
What are best practices for designing a questionnaire for the mode I am using?
Each mode GeoPoll uses has unique guidelines around questionnaire length, question length, question types, and more. During the design stage, GeoPoll will incorporate best practices into your questionnaire and guide you on how to create a survey that is optimized for the mode you are using. For more best practices for questionnaire design, visit the GeoPoll blog.
What is the implementation process for GeoPoll surveys?
- Conception & Initiation:
- Client meeting and discussion of the project
- GeoPoll internal feasibility: As described here, GeoPoll’s team will review project specifications and determine feasibility and pricing guidance
- Budget discussions: During this phase GeoPoll discusses sampling frame, data deliverables, and other items that impact budget.
- Contract signed: GeoPoll cannot start working on a project design or implementation until we have a signed contract in place. Once we have this, the project will be handed off from your sales representative to your Project Manager, who will oversee the project implementation.
- Definition & Planning:
- Survey and questionnaire design: GeoPoll can assist with survey and questionnaire design. Once a questionnaire has been finalized by the client, we will script it and send a test survey to the client for final approval. The length this takes varies widely based on how much the questionnaire needs to be built out and the number of rounds of revisions it takes to finalize a questionnaire.
- Interviewer Training: Voice call and face-to-face interviews require interviewer training. GeoPoll has a robust training program in place, and for each new project interviewers undergo 2-3 days of training on calling procedures and ethics, best practices for conducting interviews, and the specific project objectives and questionnaire, followed by mock interviews which are overseen by the call center manager.
- Survey Pilot: Many of GeoPoll’s projects undergo a pilot with a smaller sample size to ensure the questionnaire is performing as expected. Results from this pilot are analyzed by GeoPoll and can be sent to the client for additional analysis before moving forward.
- Launch & Execution:
- Survey Production: GeoPoll then moves a survey into production. For SMS and web link projects, GeoPoll’s system will select a number of respondents to be sent a survey invitation. The number of respondents who are sent the initial opt-in varies based on typical response rates in the country, incidence rate, sample size, questionnaire length, and other factors. GeoPoll then continuously monitors the sample performance, sending reminder messages and additional survey invitations incrementally until the desired sample is achieved. For voice call and face-to-face research, interviewers are given a list of phone numbers or an area to start in and begin making calls until the desired sample is achieved.
- Performance & Monitoring
- Quality Checks: During the production phase and after data collection has finished, GeoPoll’s team undertakes multiple quality checks to ensure there are no issues with the questionnaire or sample, examine individual respondents for flatlining or other unusual response patterns, and identify any other unexpected behavior or challenges during data collection. If challenges do arise, GeoPoll may pause data collection to fix any issues.
- Interim Data Delivery: GeoPoll reviews interim data during survey production and if requested can provide this data to the client for additional analysis.
- Output & closing
- Data Analysis: Once data has been collected and QC’d, it is passed onto the GeoPoll research team for analysis. The research team will do another quality check of the data and then prepare it for the appropriate data output.
- Data Output: GeoPoll prepares analyzed data in the agreed-upon output, which can range from raw data to tables and cross-tabulations to full written reports.
What sort of response rates do you normally see for mobile surveys?
GeoPoll has run extensive A/B experimentation to optimize the survey response rate across all demographics and countries, and has found that response rates vary widely based on target population, country, survey mode, topic, and more. During the feasibility process, GeoPoll can share expected response and completion rates for your project specifications, and GeoPoll’s project team will be responsible for ensuring the highest possible completion rates. Additionally, research by Lau et. al. 2018 indicate that higher or lower response rates do not necessarily dictate a more ‘valid’ survey when looking at mobile surveys in emerging markets.
What incentives does GeoPoll provide for survey participation?
GeoPoll offers a modest incentive for respondents who complete surveys. This incentive is typically less than $0.50USD equivalent in local currency deposited into their airtime account that they can then use to purchase services with their mobile network operator such as voice calls, SMS, or web data. GeoPoll can also offer other rewards systems for respondents with post-paid, or contract, mobile phone plans for whom airtime credit is less of an incentive.
Do respondents have to pay airtime credit to take surveys? What if they have a zero balance?
In countries where GeoPoll is integrated with Mobile Network Operators (the majority of countries we work in), SMS surveys are sent via a zero-rated shortcode, which means they are free for the respondent to receive and respond to, and can be answered even if the respondent has no airtime. Voice call surveys are initiated from GeoPoll call centers and do not cost the respondent to answer. Mobile web and mobile application surveys may require the use of respondent airtime for data, but GeoPoll aims to account for this with the incentives offered for these surveys.
Quality Control and Data Delivery
How does GeoPoll assure quality at all stages of the survey process?
GeoPoll’s research and operations teams take multiple steps to ensure the highest quality data. These include working on questionnaire design with clients to ensure questions are optimized for the selected mode, testing surveys with local partners, and conducting full pilots.
In addition, the GeoPoll system provides automatic verification of unmatched responses (those not within the expected response range), and upon survey completion the GeoPoll team reviews data, checking for unusual survey response patterns, drop-offs on specific questions or potential skews in the data. GeoPoll diligently monitors and regularly reviews its standardized data processing procedures, both automatic and human, to ensure that data is clean and high-quality.
GeoPoll is primarily responsible for data quality checks and clients are encouraged to raise any issues they might find or concerns they might have. GeoPoll has several data quality checks in place. These include:
- Automatic system validation and coding of responses
- Checks on satisficing and straight-lining
- Checks on contradictory answers
- Checks on respondent consistency through repeated follow-up demographic surveying
Are you certified to any specific quality system? If so, which one(s)?
We are members of the European Society for Opinion and Marketing Research (ESOMAR), World Association of Public Opinion Research (WAPOR), and other local research boards, and comply with those best practices in market research.We do not have any formal quality system certifications and are open for any client to review our processes and procedures to ensure they are up to their standards.
How long does data delivery take?
GeoPoll prides itself in our speed of data collection and delivery, however exact timelines vary based on mode, respondent targeting, the format of data delivery, and the speed with which the questionnaire is reviewed and approved by the client. SMS, mobile web, and mobile application survey results can be delivered more quickly than voice call or face-to-face interviews, which require additional steps to complete.
In what formats does GeoPoll deliver data?
GeoPoll can deliver data via multiple formats, including raw data, SPSS, Excel with tables and/or graphs, full written reports, presentations, infographics, or interactive online dashboards. GeoPoll also has APIs which can be used to ingest data directly. The data delivery format is decided during the contracting phase as more in-depth analysis will have an impact on pricing and delivery time. This blog post details our most popular data output options, and GeoPoll’s team can provide additional examples on request.
How is GeoPoll assisting organizations to gather data during COVID-19?
GeoPoll’s mobile methodology and existing connectivity to mobile network operators around the globe has allowed us to assist humanitarian and private sector organizations to gather data safely during COVID-19. Since March 2020, we have conducted extensive research on the direct impacts of COVID-19 on food security, agriculture, and other topics. We have also aided many organizations in quickly transitioning their existing research projects to remote modes that provide them with quality data even when they are unable to conduct in-person research.
In addition to the work we are conducting with clients, GeoPoll has conducted several studies ourselves, and the data from these studies is free and available for anyone to use with attribution to GeoPoll. Data GeoPoll has gathered can be found on this page, and you can email [email protected] for raw data or other questions.