Scaling success with AI in market research

Insight experts have been steadily adopting artificial intelligence-based automated software to ease the more mundane tasks off their plate, with newsletter scheduling and other communication aspects now so commonly automated it seems almost odd when they are not. This has had the impact of increasing the speed of research and team efficiency, bringing it more into line with stakeholder expectations and the ever-increasing speed of business. But it seems that there is more that artificial intelligence can offer in this regard in the present day than ever before.

1.       Informing Research Design

Research design has been a significant time-consuming process in the market research experience, with collaborations between insight experts and stakeholders resulting in exceptionally tailored research projects that generate incredible insights. It is where the creative skills start to come into their own, but it is also when we might fall back on old designs that we know have worked well in the past. It is an essential part of the process, and now there are AI-tools such as ChatGPT and Google’s Bard application that can help us generate new ideas and create new templates for research tasks in the design stage.

While we still need insight experts to build the content calendars, programme tasks and schedule them to go live, we would be remiss if we did not use AI-based tools and programmes to start the ball rolling so to speak – to give suggestions for research questions, task ideas, and base copy for community rules, all for insight teams to then edit to fit or disregard in favour of another they might think is better suited. Whether we take these AI-based ideas and run with them, we use them as inspiration for something better, or we understand why they will not work use them to figure out the right path – AI tools are here to help with research design in whatever way they can.

2.       Innovative Qualitative Analysis

With the rapid advancements of artificial intelligence in all industries, it is about time we see some advancements in some more time-consuming areas of the research experience. Areas such as data analysis. There are many quantitative tools available to insight experts to sift through statistical data and present them in graphs, tables and charts – but only very recently have we seen some tools appearing on research marketplaces that work with qualitative data.

With new evolutions of artificial intelligence such as generative AI, there is renewed potential for insight teams to hand over the initial analysis of qualitative data to automation. Generative AI tools like FlexMR’s TextMR promises fantastic automated text analysis, not just through keywords, but sentiment analysis with thematic tagging, which helps insight teams jump to the most important part of analysis – high-quality insight generation.

The time saved from tools like TextMR means insight experts can act more like consultants, focus on getting stakeholders to activate the insights they generate and make use of them throughout an organisation. Qualitative analysis especially has been a time sink, previously necessary for good qualitative insights, but now we can start from further down the race track and communicate qualitative insights faster than ever before.

3.       Enhanced Communications

Communicating with participants, whether that is through scheduled research tasks, through regular updates and newsletters, or simply ad-hoc replying to participant questions and concerns, could be considered a full-time job. In fact, there are engagement managers who are solely responsible for this role in an insight team; but not all insight teams can afford to dedicate someone to this task and this part of the participant experience suffers for it. This is where automations and AI tools come in.

Tools such as ChatGPT for participant communication could be used for some basic idea and text generation for researchers to then edit with brand tone of voice and additional context, as well as using applications like Hemingway Editor or Grammarly for any proofing needs; but there are also more automation tools available around scheduling communications and providing initial responses to participant queries. Chatbots to categorise and determine the priority of a participant query could e incredibly useful for researchers who are strapped for time, and email scheduling is something most email platforms can do too.

But participant relationship management can also be managed in-house from some research platforms such as FlexMR’s InsightHub. Alongside integrated data collection and analysis tools is an area to create impactful newsletters and send them to pre-constructed consumer groups to close the feedback loop. Other participant communication can be accessed through the direct inbox on each participant and researcher accounts, or through moderation on asynchronous research tasks such as focus groups and surveys.

Further Embedding AI in Market Research

Whilst we clamour about the evolving potential and present-day application of artificial intelligence, we must also consider the publicised warnings surrounding its adoption into all aspects of the research process.

There will always be a need for insight experts in all industries no matter how far advanced artificial intelligence comes along. We must always check it is work over, and use it as a base rather than the final product; there will still be some work to be done in these areas, but we can tackle them faster with a base there to edit first rather than having to create it all from scratch.

AI does not know how to design research that is fair and empowering. It does not have innate contextual knowledge like researchers do. So, while we can use AI to help us along and there will undoubtedly be more artificial intelligence software developed in the very near future for us to incorporate into other aspects of the research experience, there is still a job there for insight experts in businesses across all industries.