Innovation in Consumer Insights: It MattersGregg Archibald, Gen2Advisors
Innovation needs to take place where it hurts most. Gregg Archibald from Gen2Advisors, located in the US, exemplifies four pain points and their implications for the industry.
Innovation in the insight industry is the key to growth of this industry and probably the only key. As we look at the struggles for “traditional” firms and some of the successes in our industry over the past few years, it is easy to see that innovation in technology, methodologies, data sources and integration, business models are improving our overall success and impact.
But not all innovations are created equal. Innovation must noticeably improve upon some point of pain in the current process. And there are many. It would be great to put these in order of importance, but that is just impossible as different ones are important to different organizations for different reasons. So this will be a list of some the most notable pain points – and areas of innovation that are reducing that pain.
1. The speed of business
The speed of business is often faster than the speed of insights. Speed has always been an issue whenever consumer insights is involved. The issue is not just the speed of the research process – including design and analysis. There are others that are equally important such as the time it takes for a business team to get the research team up to speed, the time it takes for the research team to define the recommended action steps back to the business team, and then the time to act on those recommendations. Innovations that meaningfully reduce the time friction tend to be successful. Let’s look at a few of these.
We always need to reduce the time in the study process. From tweaking question wording and anguishing over the attribute list to coding the open-ends, looking for meaning in the tabs, or going through the focus group tapes to moving this chart or table into the powerpoint deck – the process is be cumbersome and time-consuming. At the same time, companies are bringing products and services to market ever faster due to agile processes, the rise of digital products and services, and the need to stay competitive. Innovations are reducing the time it takes to go through the research processes and keep up with the rest of the organization. There are several successful innovations that reduce the time (and cost) of the research process.
• Templated studies that are the 90 percent solution and are available on a number of platforms.
• Text and video analytics that use artificial intelligence, Natural Language Programming, or even Boolean logic to speed the process of qualitative research.
• Reporting tools such as dashboards and MS Office integrations reduce overall processing time for reporting.
We expect to see many more innovations in these areas due to the continual improvements in artificial intelligence to identify meaningful or differentiated learnings from data.
When the process of getting information to address a business question is easy enough, the business leader will often bypass the research department. The benefit of bypassing the research department is usually to reduce the time it takes to get an answer (sometimes the cost). By far, the most prominent example of this is Survey Monkey. Survey Monkey is valued at more than $1 Billion USD, not because market researchers loved the tool – in fact, many hated it. It is valued at that because it puts the information quickly (and cheaply) in the hands of people that needed to make a decision. Qualtrics is the other unicorn in our industry – and their success is largely for a similar reason. There are many other examples of insight tools that make it easy for the person needing information to get that information. Providing valuable information to the end user while streamlining the internal business process is a near perfect recipe for success.
2. Limited perspective on a problem
Another pain point in the research process is the limited perspective a research study often offers to shine a light on a business problem. This is most easily illustrated by a customer satisfaction example. Most satisfaction studies utilize surveys to gather information from a consumer based on the use of a brand or triggered by a transaction. The survey covers anywhere from two questions to two hundred questions (I’ve seen them!). But that consumer information is only part of the information needed to determine action steps to address customer satisfaction problems. There is other data that could be important to understanding the problem - sales data, CRM systems, logistics/channel information, competitive intelligence, and even weather data. Methodologies and tools that allow a problem to be understood more holistically are certainly an area where innovation can and is making a big impact. Many high growth companies are acting as DMPs to improve the cross-data insights that can be delivered to address the business problem more holistically.
We are seeing successful innovation in this area across business issues and research processes. Several companies are providing brand health programs that integrate internal data (primarily from CRM systems) with external primary data coming from a variety of sources including advertising data, social data, ratings and reviews, competitive sales data, along with attitudinal data for a much more comprehensive view of the business issue. Attribution models are continuing to improve our understanding of advertising ROI by considering hundreds or thousands of variables at the same time. These are only a couple of examples.
From a research process point of view, advancements in data synthesis and predictive analytics are both improving the capability to see the business world more holistically – at least behaviorally. This area is still in its’ infancy and we are expecting more and significant improvements in this area. The growth in usage of Knowledge Management Systems bodes well for the infrastructure for these innovations as more consumer insights are stored in a single access point.
3. Data quality
Data quality is another significant pain point and another area primed for innovation. The term “data quality” is used here with its largest possible meaning. This is not just about sample quality, though that is part of the story. This is also about the accuracy and completeness of data on any single person – across behaviors, sectors, channels, time, and life stage. This is way more than we can ask, and way more than a person can answer, and way more than we pay that person for. Innovation in this area has been relatively limited, but expectations are high. Most innovation has focused on digitally tracked behaviors. Passive data collection through the mobile phone can track some geo-location data information and media exposure. And we have seen some innovation in tracking sales information of an individual across stores and websites. The scale of these efforts is still limited. Single source data has been a desired tool for many years (and a solution to sample quality issues). Single sources allow a very broad perspective of a person’s behaviors, attitudes, influences, and motivations. Blockchain technology offers a very promising solution to this this as the data is of almost guaranteed quality, the person can be reasonably compensated for sharing their data, and the data has the ability to be tracked across behaviors, sectors, channels, time, and life stage. Innovation in this area is going to transform insights as a foundation for problem definition, access, new analytic methods, and new perspectives on the nature of behavior and choice.
4. What's the human truth?
The final pain point to be discussed is achieving an understanding of the human truth. To date, we have made progress in our frameworks and tools for discovering “truth” from where marketing research started in the 1920s and 1930s. I would argue that two of the most significant advancements were the application of conjoint analysis to marketing issues in the 1970s and the application of behavioral economic models to consumer insights (mostly in this century). The reason behind this is that both are using context to understand outcomes. It is the context of features for conjoint and the context of emotions for behavioral economics.
However, the information, tools, and techniques that we have to understand behaviors, motivations, and the truth surrounding those, in the context of a single decision, are limited. Let me explain by example. Researchers in the life insurance industry know much about a consumer’s path to purchase once a decision has been made to buy life insurance but know little about the decision to buy life insurance in the first place. In the beginning of that decision, it is about the opportunity cost compared to buying a new couch or going out to dinner or any of a million other things that can be done with the money rather than the features of one product compared to another. It is that initial alternative decision that must be made in the context of a person’s entire life.
Behavioral economics has led us to the role of emotion in a choice, but our models have done little to integrate those emotions in a single or a changing context. The emotional context behind the way I feel about my significant other and my best friend are very different, even in the same conversation. If I choose to treat each the same, I might well soon find myself with neither. The same logic applies if discovering a new wine for a nice drink on the back porch versus a dinner party. New platforms and techniques that improve our ability to capture information relevant to the question of emotions, and hence, values and motivations will improve our ability to understand moments in the context from which they occur. This is just the tip of the iceberg. As the data made available by a true single source foundation and the future capabilities to be exploited from deep learning become more available, our ability to truly understand the complexity and nuance of human behavior will be one of the most important drivers of innovation.
I do not wish to imply that all of the pain points and innovations are discrete unto themselves. Some of the innovation is based on addressing two or more at the same time. Nor do they completely eliminate the pain points – which is why innovation is a long and incremental process. But progress is good.
Of course, these are only a few of the pain points and therefore, drivers of innovation. These are certainly some of the most pervasive. Others that are important include the cost of research, types and sources of data, ROI-based evaluations, and tools to improve our ability to influence. There are innovations in all these areas that make the job of marketing research easier and more impactful.
Gregg Archibald is the Managing Partner for Gen2Advisors, a consulting company dedicated to the insights industry. Gen2 works with both brand research departments and research suppliers, giving Gen2 a unique perspective on the industry.