Which insight suppliers and clients are the most innovative? | READ THE LATEST GRIT REPORT

10 Ways Sample Companies Should be Using Automation

Sample automation is a great potential tool for improving survey processes. Many opportunities exist for dramatic improvement, and the industry needs to embrace them for the opportunity to be realized quickly.

Editor’s IntroSurvey research is increasingly challenged to be better, faster and cheaper. Automation of sample is one important tool to help the industry get there. JD provides a great overview of where we are with sample automation, argues that it is underutilized, and gives a vision of how the industry can do much better than it is right now. This is a timely, important article for survey researchers to read and take action on.

The insights industry loves to talk about automation, which has the potential to speed up processes across the board. In the sample space, current applications are often limited and difficult for buyers to understand what suppliers are actually doing and what benefits are on the table. In fact, it is precisely because most applications of automation are limited that people feel it is more buzz than brilliance. But the truth of the matter is that, when used to its fullest potential, automation can have massive benefits for process and outcomes.

Sample companies can be using automation throughout both their internal and external processes in order to create not only efficiency and lower costs, but also fundamentally better data and agility for clients. Here are 10 ways that sample companies should be using automation:

1. Respondent Engagement

This application of automation is barely used in the industry, yet it represents a concrete and necessary way in which suppliers must take responsibility for sample quality. Sample companies have the ability to track respondent touchpoints ranging from recruitment source, to profile data, to project experiences. These data points can help to indicate the level of attentiveness and engagement of the respondent, helping to determine if the respondent is “real.” Even surveys can be evaluated in this manner, using data such as field statistics and respondent ratings. Automating the evaluation of respondent engagement and survey quality up front can help to not only match the right person with the right survey and maximize the likelihood of survey completes, but it can also have a positive impact on things like over-quota, recontact and retention rates. Taking a thoughtful automated approach to respondent engagement can lead to significantly higher satisfaction and engagement. That means better data.

2. APIs

It is important to remember that automation is not the same thing as an API, or automated programming interface. An API is the exchange of structured information between two systems or platforms. At a minimum, sample APIs facilitate the exchange of information on project specifications. Sample companies should be well-versed in custom APIs to programmatically exchange project specifications to quotas to demographics and hundreds of profile points between supply and demand partners.

3. Buying & Selling

Buying and selling of sample is the most widely-used aspect of automation in sampling. Here, buyers and sellers share project specs and sample counts via an API. At the point of project commissioning, sellers launch sample at the project and transmit demographic information about the respondent. What gets transmitted is often limited though. The process of mapping data fields—that is, sharing demographic and behavioral data about respondents using pre-arranged codes—is tedious and requires regular maintenance. When it is not done, respondents “leak” out of the system or have to repeatedly answer questions to which suppliers already have answers.

4. Feasibility

Over the past few years, the feasibility process has changed dramatically. Yet, many sample providers still manually produce feasibility calculations in spreadsheets using simplistic formulas. Automated feasibility processes can not only take into account a huge number of factors–from field parameters to individual behaviors–to provide accurate estimates of how a project will fill, but also deliver greater speed, accuracy and dependability.

5. Quota Management

Automation allows us to keep a silent and constant vigil on each and every project, detecting anomalies instantly, closing quotas automatically as they fill, and otherwise quickly and efficiently ensuring projects remain on track and raising alarms when they are not. Automation thus brings peace of mind to the research buyer and better experiences to the respondent. In a manual environment, problems usually aren’t detected until it’s too late, and respondents suffer.

6. Recruitment

Recruitment is one of the areas that suppliers have at least partially automated, but rarely to its fullest effect. When fully automated, sample suppliers can trace panelists all the way back to the source to optimize sourcing based on quality scores, which intrinsically ensures that they are recruiting real, attentive, engaged, deeply profiled people who genuinely complete studies. By using automation effectively, sample companies can allow for on-the-fly recruitment based on need. Automating the entire process allows the management of more recruitment partners, which means greater reach and diversity, and less “groupthink”. When automation is properly implemented in the recruitment process, it improves everything else downstream, from feasibility estimates to response times to economics for suppliers and buyers.

7. Field Monitoring

Field management is perfectly suited for automation. Very similar to #5 above, Quota Management, sample companies should be using automation to achieve the same goal: making sure nothing goes wrong in field immediately. This nimble and fast approach means problems are less likely to slip through the cracks and can be addressed before they snowball.

8. Incentive Management

Instant rewards is the name of the game when it comes to respondent incentives. Automation of incentives allows immediate delivery, closing the loop with the respondent. This is another place where most suppliers are already using automation.

9. Fraud Mitigation

When it comes to fraud mitigation, there is much work to do to combat the growing menace of fraud. Traditional techniques are still in widespread use, but need to be coupled with more technologically advanced methods to fight the savvy fraudster. Using artificial intelligence to navigate billions of data points that include demographics, behavioral indicators and field experience can make great strides in detecting and combating fraud. A smart machine can instantly analyze anomalies and patterns in the data, such as user “surges” involving specific demographics, IP ranges or completion time frames. By using this kind of technology, sample companies can start to significantly mitigate fraud and meet new threats.

10. Profiling

Automation at this stage can use respondent profile data to enable precise sample targeting across hundreds of qualifiers. This can ensure that the right respondents are matched with the right study without the annoying requirement (for the respondent) of having to answer the same questions constantly! Automation can help ensure proper targeting, improve the user experience and yield better, more accurate data.


We hope this article has helped you understand two things. One is the real potential of automation in sampling. The other is that this potential is already within reach and utterly essential for suppliers to implement. It’s time for our industry to start using automation to its fullest. Instead of partial implementations that create broken processes, using this technology to its full potential can help to build true efficiencies that result in better quality from start to finish.

Please share...

Join the conversation