12 Key Considerations When Choosing a Mobile CAPI Vendor

As more and more market research firms transition from PAPI paper-based surveys to mobile devices (such as tablets or smartphones) for offline CAPI, multiple vendors have started offering solutions. Due to the inherent complexity of field work, there are huge risks of data loss, quality control issues and sub-par technologies that can severely limit your ability to successfully execute a specific project. And there are many risks to data quality and integrity for the entire project. In this article, Dooblo summarizes the 12 most important considerations when choosing a mobile CAPI vendor – so you can avoid the pitfalls and find the best solution for you.


#1 Quality Control Capabilities

Everybody knows quality of data is key when working in the field. Today, with mobile devices, things got a whole lot better. It is, therefore, critical that the CAPI vendor allows you to perform tight quality control and assessments on the interviews and the interviewers work, such as GPS location capturing, silent recordings and silent photo capture are all critical for top notch quality control. Look for Predictive Quality Control –script codes that spot quality control issues in real-time and while the interview is being performed; these are the next generation of quality control and a must-have when scaling your QC operation.



#2 Dashboards & KPI’s of Fieldwork Operations

One of the key aspects of switching from pen & paper to mobile devices is to make the field operations more productive. To achieve this, it is critical that the solution includes key fieldwork operations dashboards and reports. The CAPI software should therefore include for each project a set of dashboards that will provide fieldwork management with key performance indicators about the fieldwork. As the famous saying goes: “what you don’t measure you can’t manage”.



#3 Support for Super-Complex Scripting:

Face-to-face projects tend to go hand-in-hand with very long and complex questionnaires. Things like Kish-grids, LSM (living standard measure) calculations, respondent segmentation logic, complex tracking sheets and conjoint studies are all very common requirements for face-to-face surveys. Moreover, face-to-face questionnaires tend to be 30-90 minutes long and include almost all question types you can think of. Therefore, it’s critical the CAPI solution supports at least the following question types: standard questions (like single choice, multiple selection, open-ended, scale, numeric etc.), dummy questions and all forms of grids. In addition, as the project will be conducted in the field, it should also support: signature capture, sound recording, photo and video capture questions, GPS Location & offline mapping, bar code & QR code capture, Counters and External List questions. On top of that, the CAPI software should support advanced scripting logic such as skipping, piping, branching, looping, filtering and randomizations of both answers and questions and entire sections of the questionnaire.


The script editor should support auto-complete of scripting functions and the scripting model should allow the scripter to write code that can access and interact with questions, answers, dummy questions, global variables etc., be able to accept external data lists and run a simulator. Finally, the CAPI solution should allow the scripter to quickly generate test (dummy) data to both check that the logic works but also send off to data processing to confirm data structures BEFORE going to the field.



#4 Offline Data Collection

The ability to collect data while offline is absolutely critical, so the data collection app must continue to function normally in offline situations. The app should support the following operations when offline: completing full interviews and if the survey contains any videos or photos or recordings, these should be available while offline. The mobile device should allow the last user that used the mobile device to log in even if the device is offline and of course that the data should be automatically synchronized when the network is back online.



#5 Data Output

Market Research firms use a variety of analytic applications to analyze data. To make the data processing departments more productive, the solution should allow exporting the data to the native file formats of the most common tools used in the market. The minimum list of data formats that the CAPI solution should export to is: SPSS SAV files, Triple-S, ASCII, CSV, Microsoft Excel, Microsoft Access. In addition, as many global MR companies use Dimensions (now called Unicom Intelligence) and Quantum, supporting native export to the Dimensions DDF/MDD file formats and Quantum axis and data files is highly recommended.



#6   Multi-Language Support

Many face-to-face projects run in multiple countries or in different areas in the same country, and so the same questionnaires need to be run in different languages. In the field, switching between the available languages should be super-easy to do to facilitate interviewer productivity. Thus, instead of generating multiple copies of the same script, it is vital the CAPI solution supports both adding multiple languages to the same script and easy importing of the translated files.



#7 GPS Support

Today, GPS tracking and reporting is one of the de facto pillars of quality control, allowing field/QC departments to quickly verify that interviews were performed at the correct locations and that interviewers were covering the correct areas. The CAPI solution should both capture GPS information but also include a back-office screen that shows the interview locations and walking routes of interviewers between date ranges and display the information on a map.
As GPS is so critical to quality control, the data collected should also be exportable in the data output file as part of the interviews and be included in the KPIs/Dashboards of the CAPI solution.



#8 Quota Support

Almost all face-to-face projects include quotas on demographic and other variables. This means the CAPI solution should include built-in support for not just simple quotas but also more complicated multiple levels of quotas (interlocked quotas). In addition to just defining the quotas, the CAPI software should allow setting the targets per interviewer or globally per project and of course enforce these quotas on the actual devices so that the MR firm will not be under/over quota at the end of the project. Last but not least, the CAPI solution should have a quota-progress screen to allow field operations to closely monitor the progress of the quotas.



#9 Integration with 3rd Party Systems – API

Many times, customers need to integrate the solution with either in-house dashboarding/database systems or other legacy systems or just create replication of the data on the customer network and the CAPI solution must support it. The software should include both a command line tool for easy integrations and a more robust REST based API for more complex integration use cases.



#10 Solution Security

Data security is one of the most important aspects of running projects for end customers. The CAPI software should have security integrated to the core of the solution and has to be implemented in all the components of the system. Some or all of the data collected can be sensitive, therefore it is critical that the DCA (Data Collection App) has the ability to encrypt the transport of data between all the components of the solution. Other than the industry standards for server and data security, the vendor must allow customers to configure security access to data only to specific employees and only to a specific project using a role-based security model. It is important that every interview should upload to a specific “Project Area” on the data center that with access to only relevant personnel.



#11 Server Infrastructure & Application

The Data center forms the heart of the CAPI software and is responsible for safely storing all the collected data and providing access to that data for the customer. Due to the huge growth in data size and volumes, today’s CAPI solution data centers must be hosted by one of the leading cloud providers and the servers should use a well-known industry standard data storage software solution. Backups of the server data are critical and should be done at least daily. The CAPI vendor should have a monitoring team in place with monitoring software running on the data center servers and equipment 24/7. The data center should provide a segregated or virtually segregated environment for each customer so that in no way will one customer be able to access data of a different customer. The uptime of the server datacenter should be monitored and available for customers.



#12 Vendor Service

While software and infrastructure are two very important aspects of the solution, the vendor and support service it provides is a key aspect as well. The vendor support team must handle tickets, scripting and general questions, fix software related bugs or any complicated issue that requires access to the solution source code. The vendor is required, at minimum, to provide customers with training & support materials. Configuration changes to the vendor system infrastructure and software should be authorized, logged, tested, approved, and documented in accordance with industry norms. Changes to the solution which impacts the service should be thoroughly reviewed, tested, approved, and well communicated.


We hope you’ve found these tips helpful! If you’d like to learn more about implementing a successful mobile CAPI project, why not check out our SurveyToGo solution, deployed by hundreds of MR firms around the world. Learn more by contacting us here or visiting our website: https://www.dooblo.com/ .