Editor’s note: Wale Omiyale is senior vice president of market research at Oslo, Norway-based research firm Confirmit.

For years, marketing research organizations have been under immense pressure to uncover higher-quality insights faster and at a lower cost. In general, delivering faster insights at a lower cost is not new. However, sometimes faster and cheaper can mean poor quality.

So, how can marketing researchers avoid this?

Software providers are working to refine the capabilities of automation tools used in such processes. Today, automation is paving the way to make this a possibility with accurate, agile research.

Automation can be described as the technique of making an apparatus, a process or a system operate automatically. Since computers are exceptional at following rules, software providers have used this to their advantage by programming computers to carry out information-processing tasks. If this is done successfully, users can see faster delivery and lower costs.

Automation has already been around for years in simpler forms, such as questionnaire scanning, which was revolutionary in speeding up the process of data capture and was created in direct response to the need for faster results. With this need in mind, automation tools have been reworked and redeveloped – so much so that the tools used today now span the entire life cycle of marketing research. In addition to automating survey design, sampling, data collection and reporting, tools are available for much more advanced automation techniques, from emotional response recognition to multimedia feedback, social media analysis and more.

Are researchers sacrificing quality?

With the adoption of speed comes quality concerns. Traditionally, some marketing researchers have felt that a reliance on technology could pose real challenges for data integrity. Many question the accuracy of software-interpreted data.

The concerns are valid. While the industry has yet to reach full automation maturity, there are many solutions today that successfully deliver speed, consistency and accuracy.

One example of success is in the field of text analytics. The increasing volume of social media data has been a key driver for developing automated analysis tools and social media analytics solutions are now proving to have value in delivering insight from vast data sets.

Such tools have several added advantages of paving the way for researchers to become insight experts. These include:

  • repetitive tasks associated with data collection and analysis can now be automated;
  • actionable insights and sentiment can be found within the free-form text;
  • unstructured text can increase respondents’ engagement;
  • open-ended questions can allow for a shorter survey but more meaningful, specific and useful feedback; and
  • researchers can focus on more in-depth analytical processes that require human interpretation.

In-demand skill-sets 

As automation becomes more advanced, the need for a variety of skilled people also increases. The following list provides examples of specific roles that are increasing in demand and changes in the workplace.

  • Broadly-skilled project managers: In-depth subject knowledge is being replaced with an understanding of the many automated steps of the research process. This certainly impacts the role of the research subject-matter expert but allows research teams to be more flexible in recruitment and service delivery.
  • Data scientists: With administrative tasks being handled by automation tools, many research teams are evolving into specialist hubs, where researchers become data scientists and reports become strategic business guidance.
  • Inter-departmental collaboration: While some think automation will eventually bring an end to people-based businesses, there is growing evidence to suggest that automation improves and thrives off inter-departmental working and collaboration, both across survey teams and with external partners and customers.
  • Research analysts: In situations where the program is of a low-level complexity, we are seeing the traditional research analyst taking the responsibility to self-program a study. Here, automation of functions is critical to ensure that the expanded role works effectively.

Self-serve opportunities 

Self-serve means different things to different people. At a basic level, self-serve represents the ability to run a research program with as much automation in the process as possible, eliminating the need to outsource service delivery. This could be in the simplest form of scripting a survey but ultimately extends to the analysis and report delivery of data. Automation of analysis and reporting currently work well in a standardized research approach but much is yet to be done for ad hoc projects to automate the process.

Self-serve may ring alarm bells for some, suggesting a diminishing need for the skills of marketing researchers, but this is not so. Many companies are looking for quick insight and sometimes they only want to focus on questions that get to the heart of the insight. With automation tools to support this way of working, researchers may still get 80 percent of the information they need in 25 percent of the time – which makes a justifiable business case for self-serve.

It’s important to note that self-serve does not negate the need for in-depth research programs; rather, it is a new layer that sits on top of substantial analysis and insight. The developments in automation are taking us toward a hybrid model of research, where the needs of companies are delivered in the time frames and formats most suited to each.

When implementing self-serve, it’s important to follow these steps to ensure the process runs smoothly and produces results:

  1. Weave self-serve into the company culture. For self-serve to be truly successful, it can’t be a one-off project. It must be woven into the culture of the organization through executive buy-in, incentives/rewards and identifying/addressing any obstacles up front. Usually, lines of business demand self-serve tools, bypassing the IT department which is more inclined to build from the bottom up. The explosion of self-serve supports a more autonomous working culture, enabling teams to use the right tools for the job.
  2. Don’t be afraid to keep it fluid. The beauty of IT process automation – and technology in general – is that it is continuously changing and improving. Be sure to keep an open mind and make modifications and changes wherever necessary. Before rejecting or implementing, always consider what works best for your company.
  3. Make your testing and reporting top-notch. Ensure there is always someone to evaluate the data from reports; otherwise, there is no point in generating the data in the first place. Furthermore, automation tests exactly what you tell it to – no more, no less – so be sure to incorporate manual testing as well to guarantee a complete and comprehensive assessment.

Looking at the benefits 

While there is still much more opportunity for automation in MR, it’s clear that it’s already firmly entrenched in our day-to-day processes. We must be more rigorous than ever in our assessment of quality.

Automation is used in many industries to deliver savings in time, materials and resources, as well as a tool to drive improvements in quality, accuracy and precision. By sticking to the foundations of integrity and accuracy on which the research industry is built, automated research tools can help MR evolve in the face of ever-changing client requirements.