Stephan Gans, PepsiCo & Ryan Barry, Zappi: Tech alone isn't enough to be insight-led

Consumer insights are highly influential in business decision-making, but only half of consumer insights teams are seen as strategic partners, and one in five are still perceived to be ‘order takers’.  

That’s according to research from consumer insights platform Zappi and the American Marketing Association, which surveyed 715 marketing professionals from April to June 2024 about how they manage and view their insights function. The study found that consumer insights staff within organisations were viewed as ‘strategic partners’ for 52% of respondents, ‘advisors’ for 31% and ‘order-takers’ for 17%.

The survey also found that while technology is used frequently for consumer insights, 59% reported that their insights projects are not handled systematically.

For the past few years, PepsiCo has worked with Zappi to digitise its consumer insights function and develop its ad testing. Now, the two companies have co-authored a book on the subject, The Consumer Insights Revolution.

Research Live sat down with Stephan Gans, chief consumer insights and analytics officer at PepsiCo (pictured, left), and Zappi president Ryan Barry (pictured, right) at Zappi’s recent Connected Insights Conference in New York, to discuss insights maturity and why technology alone isn't enough - people and processes are key to becoming more consumer-centric.

In the book, you outline a connected insight framework (from stage one ‘disconnected’ to stage three ‘connected’). How did you develop this?

Stephan Gans (SG): For me, it started when I started my current role at PepsiCo. I got, as one would expect, a large file of reports from consultancies who had worked with by PepsiCo to think through how PepsiCo could get more value out of consumer insights – McKinsey, BCG [Boston Consulting Group] and Bain. It was an immense file and I went through all that. There was one page of that entire file that stuck with me, which was that BCG framework.

While working one day I came up with a metaphor of the person behind the counter at the Post Office. I said, ‘Look, the guy or the woman behind the counter at the Post Office is just an order taker. He or she is never asking: ‘Hey Ryan, should you even be sending this letter?’ Or: ‘What’s in it?’ It’s just: ‘Oh, this needs to go to Paris? OK, it’s going to cost you this and it’s going to take two weeks.’ That was what market researchers in PepsiCo to a large extent just did, when I came in. The brand manager, would say: ‘I want to launch this product. Stephan, how long is it going to take to validate if this is a good idea and what’s it going to cost me?’ Then I would go to an agency and negotiate, and the whole process began and I would say: ‘Take six weeks out of your project timeline because I need those six weeks to do my project.’ And then after six weeks I would come back with a traffic light – red, yellow or green. That maybe worked 30 years ago but today, marketing is a real-time game.

Ryan Barry (RB): We originally had the BCG insights maturity matrix, and we were going to use that in the book, and I think we all felt it was good but it was a seven-year-old matrix at that point. It happened to coincide with us having a new CMO and she suggested we should do something different.

In the book there’s a framing of how Stephan internally positioned going from mail counter to strategic partner, and what we were trying to do with the framework was decouple or codify the components that made somebody go from: ‘Where’s it going?’ To ‘Are you sure? We might want to go there.’ And that was really around talent – the ways they worked with other teams and the technology and systems they used. Because even in the early days of our partnership, which has always been quite advanced, even though technology was enabling the answers quicker, the workflow didn't change. We spent a lot of time thinking about talent, and PepsiCo’s a very distributed business, so how could we get people within the businesses integrated? How do we talk to our agency partners differently? We step-changed through that. I think what BCG did was a really good job of articulating the problem but not maybe breaking it down, in today’s paradigm, where there is so much technology and change.

SG: In this framework in the book, people recognise themselves, to a rather painful extent. It’s so often painful because I have yet to find the insights leader at any company, in any industry, that doesn't aspire to have a bigger impact on the business than she or he actually has. That is the one insight into this profession that I think is just true across all industries or verticals.

RB: There was one thing that was different in PepsiCo’s business - the chief executive was actually really behind this. Often, you put people on an insights transformation journey but the chief financial officer and the chief executive aren't sponsoring it, which makes it quite hard to change. Not everybody’s going to have that paradigm but we still wanted to help people take the same steps in the spirit of a tangible book. Because most people that we need are still, with the best of intentions, trying to figure it out and it’s not a technology problem – it’s a people and a process problem.

There has been discussion in the industry for years around agencies becoming strategic partners or client-side insight professionals being strategic insight partners to the business, but it’s often not happening in real terms. Why not?

SG: The thing is, you cannot become a strategic partner to the businesses if you outsource the core of the work, which is what a lot of insights people still do. What this new way of working enforces is that you as the insights leader, as the research person – you own the data, there’s no delegation to an agency any more. Yes, there’s a tech partner and in this case, our tech partner is Zappi, but the data is on your laptop. If somebody has a question, you're able to come up with the answer in either five minutes or 50 minutes, depending on your proficiency with the platform or your ability to get hold of somebody from customer success. If you digitise a function as we did and you work with a digital partner, you get to meet a whole new type of function – called customer success. Kantar, Ipsos or Nielsen, they don't have customer success. They do the opposite – they black-box stuff. Customer success teams, if they deliver, allow people to deliver on the challenge at hand themselves. Tech companies have customer success functions, research companies do not.

So, really, that’s a technology sector approach.

SG: That’s a tech sector approach.

RB: The customer success [function] isn't saying: ‘This is what you should do with your ad,’ it's: ‘Here are six tools that you can use to figure the answer out for yourself'. I collaborate with some of the post-secondary education partners in this space. They're teaching people how to be analysts, how to do statistics and how to do sample frame, but they weren't teaching people customer success. If you're going to go brand-side you need storytelling skills, business acumen, critical thinking principles. If you're going to work agency-side and you're just out of a master’s degree, the best place to start is actually customer success.

We have six alum from Michigan State’s master’s programme but one of the things I've been asking [the programme] to do is have a customer success curriculum that teaches folks just enough to be dangerous and how to use software to do their jobs. I think there’s a change afoot on this topic from an education standpoint for a lot of people coming into the space, which is good for the industry on a net basis. Three of the people that work on our PepsiCo account weren't researchers. Pooja [Bailey], one of our customer success managers, ran a Marriott before this, so she had skills of hospitality, patience and teaching people.

That works on both the corporate side and our side – how do we get people who aren't thinking about this space into it? Because it is a very niche industry. I think customer success is a wonderful way in.

SG: I'm not sure I agree with niche; it’s a super conservative industry. Why? Because people always have been pushed into roles like: ‘Hey, the buck stops with you. I'm handing you my baby for six weeks and you're going to tell me if it’s a red, yellow or green.’ That’s a lot of pressure, so people don't want to mess that up and everything becomes, in a way, conflict-avoidant and conservative as a result.

Ryan, you mentioned during your presentation that you’re having conversations with chief data officers and on the whole they are not integrating customer data. What needs to be done to address that?

RB: Digital transformation’s been around for a long time, but big businesses still have seven different accounting systems, let alone insights. So, there’s a basic integration of everything that’s operational anyway, so that’s the first point.

Secondly, there is transformation in adjacent industries. The media industry became programmatic while our industry was figuring out whether we could even use technology. So, the world became more divided and now you're in a place where the big creative networks are creating content on the fly using language models. So, because we weren't intentional over the systems that we used in consumer insights, data doesn't exist in a place where most brands own it. And even if they do, it’s not being used and collected. When we first met, PepsiCo had 30 different ways to evaluate a claim across the business. If you're not intentional with how you do primary day-to-day learning, how are you going to put it in the matrix? 

SG: A company like mine doesn't have a culture of saying: ‘I want to understand how a sports drink can appeal the best to female weekend warriors in the west of the United States. OK, we've tried to appeal to those people with ads for five years, we've made a total of 35 ads, let’s ask the system, and now the AI, to tell me the five dos and don'ts when trying to appeal to this cohort.’ That’s called a meta-learning, but it’s just a fancy way of saying: ‘Let’s do your homework'. But when you digitalise the process and you standardise the tools, suddenly you're creating this wealth of skill that you can leverage.

RB: That concept, it’s very exciting to me not just because of the data integration but we get opinions from humans, and there’s a lot more ways to monetise your attention span now than there was when we started surveying people on the internet, and there’s a quality problem in this space because of this reality. So, if we start with what Stephan said, it's, ‘OK, what do I know? What are the dos and don'ts?’ Then I can afford to leverage that and ask people two or three questions, which means I can ask them differently, I can have them record voice or video. If we play this right, the robustness of our data becomes richer because we're going to be able to have more robust human conversations than a 22-question typing tool.

Even today, we still ask the same questions of people over and over – why? We already know where you're from, your age, where you shop, your preferences, and in the case of our work together we know how you respond to advertising for different categories.’ So, maybe we could just test the new assumption that we don't know the answer to now and then keep evolving that? And that to me, I think it’s actually quite existential because I don't know that you get around the data quality problem without changing the way you interact with humans.

If there’s one key thing that companies can do to be more consumer-centric, what would it be?

SG: It comes down to empowering the people in the room. So, I'd say that it’s driving consumer-centricity at the point of commercial decision-making, that needs to happen, and getting the insights person at the table when there’s commercial decision-making happening, whatever the decision is, that’s one step. But empowering that person to really bring the potential to leverage the wealth of data for insights and added value, enhanced competitive advantage on the spot – that is progress. The way I like to talk about it is global might for the local fight – in a company like mine, which is like the Philippines, like thousands of islands and everybody’s doing their own thing. [It’s about] creating that federal structure that enables all these islands to thrive, leveraging the skill for advantage, that is what you get if you start working that way.

RB: I often draw parallels, ironically, to the SaaS industry. When CRMs and software platforms became a thing, there were all these budding technologies created with one goal in mind – to enable sales and marketing teams to be more productive. But the result was they were just buying technology and throwing it into the existing ecosystem and nothing was changing – sellers and marketers were less effective. Then the dawn of revenue operations became a thing, and you will not find a software company that doesn't have a pretty robust RevOps team curating the tech stack and thinking about processes.

To me, the shift is moving to a system like that where those who used to evaluate copy-testing vendors are responsible for curating the ecosystem that works locally and not just sitting in New York City or the Netherlands, but actually creating a system that works locally but having the global might. I think the only way you can do that is to empower teams with a mandate from the CEO, in their earnings report, saying they are customer-centric and actually enabling the ecosystem. I always draw the RevOps parallel – it was a mess 10 years ago because everybody had 50 tools. We are in a space with a lot of novel methods, a lot of really interesting ways to learn from consumers – there are something like 1,000 platforms on the Insights Platforms site. But how does it actually all work together? How does that then free up humans to be empowered, to be strategic?

That’s a lot of what this book is: how do you audit what you're doing, the people you've got, the way that they work with adjacent functions? And then what are the steps you can take?

SG: It’s one thing for people to recognise where they are on the framework but of course, closely linked to that, it’s something completely different to think about whether this role is actually for them. A lot of people are completely happy and comfortable in more of a project management type of role. But this throws you in at the deep-end, this empowers you and ultimately raises the bar of what people expect from the value that you bring. So it’s not for everybody and that is totally fine.

RB: But that also can really help open the talent pool. Often, good market research managers with a project management background get promoted into this job and then are asked to transform a function – those are completely different skill sets. I'm not saying those folks can't evolve but I think the enablement of software allows us to attract a different brain. One who can actually sit with a CMO and say: ‘You're completely off base here, we should do this’ and that person will actually listen to them.

SG: Nine out of 10 times, the best surgeon gets made head of surgery. But the best surgeon enjoys doing surgery and that’s what she needs to be doing, and somebody else probably gets a lot more energy from making the weekly schedule or negotiating the budget or setting the vision for the surgery team, right? These are complementary skills.

‘The Consumer Insights Revolution: Transforming market research for competitive advantage’, authored by Ryan Barry, Stephan Gans, Kate Schardt and Steve Phillips, is now published.

This interview has been edited for clarity.