Solutions

We tackled every challenge with a futuristic solution so our client could stay ahead in the game:

• 01. Integrating Fraud Prevention Services
  We used a mix of third-party integrations and in-built algorithms to build several layers in our fraud detection mechanism.
  Fingerprinting Method: We enabled user IP tracking and PID tracking along with any related survey studies taken in the past by the user. A the researcher can choose all three or any of these fingerprinting methods to permit unique users or respond-ents to take a survey.
  External Fraud Detection Services: We integrated fraud detection services like Relevantid and SampleChain for an additional check to ensure we redi-rect only genuine respondents to the survey site.
  Open Ended Validation: We implemented an OE validator for advance verification. Selecting this option fetches a couple of random questions from an in-built question library and are presented to each respondent. Answers to these questions are then ana- lysed for parameters like context, slow or fast response, mistakes, etc. using an R & Python algorithm. Based on this analysis a score is allocated to each respondent and based on this score the respondent is either redirected to the survey or back to the supplier site.
  Length of Interaction: Furthermore, we used the approximate LOI suggested by the researcher as a benchmark to pass responses as qualified. For advanced survey hosting sites that have their in-survey behavior analytics mechanisms in place, we enabled a server-to-server call to confirm the validity of the responses after a survey has been completed by the respondent.
• 02. DIY Tool
  To quickly process the high volume of survey requests, we created a DIY tool for researchers to set up and manage their surveys. We built a feature to check the financial feasibility of the study. Based on the sample size, LOI, and demographic and geographic criteria of the target audience, the feasibility tool can provide the best IR and CPI for the survey. Once the researchers have created a survey they can select panels based on their sampling requirement. They can also restrict the number of responses from a particular panel.
• 03. Streamlined Business Processes
  To eliminate third-party dependencies and switch between the systems to manage sales to delivery cycle and billing & invoicing, we developed bidding and accounting modules in the tool.
  Bidding: Project managers can set up an account for a client and initiate a bid for each survey request received from the client. After the bid has been approved by the requesting client, the bid amount will be added to the associated account for future billing.
  Accounts: We integrated QuickBooks API into the accounts module of the tool. Once a survey status changes to complete, an invoice can be generated and sent to the client through this module without having to switch channels.
  Reports: We built the report module for the client to track monthly sales & billing reports, team load management and sup- plier performance etc. By eliminating these third-party dependencies, we saved our client 50,000 USD in licensing costs per year. Other than the above-mentioned challenges and solutions, we tackled the increasing server load by using AWS load balancers and created primary and secondary servers in Mongo DB for faster performance.

• 

  Integrating fraud detection techniques

  We enhanced fraud detection using various methods including fingerprinting, external services integration, open-ended validation, and interaction length validation.

• 

  DIY tools

  Allow researchers to manage high survey volumes efficiently, including a feasibility check and panel selection with response limits.

• 

  Streamlines business process

  We removed third-party dependencies by integrating bidding and accounting modules, allowing seamless survey management from sales to billing, with savings of $50,000 annually.