Summary

Molly Strawn-Carreño, Director of Brand Growth at aytm, a market research technology company, discusses why data alone is insufficient and how genuine consumer understanding drives better business decisions. Drawing on practical examples and professional experience, she explores the importance of listening, behavioural gaps, data literacy, B2B research, and the role of AI in modern market research .

 

Key Points

  • The Best Brands Listen Continuously: The best brands maintain constant, meaningful dialogue with their customers across platforms such as TikTok and Reddit, rather than relying on occasional studies or assumptions [1:20] [1:54].
  • The Organic Strawberries Story: A shop-along study revealed a mother buying organic strawberries for her children but non-organic broccoli for her husband, illustrating how real behaviour can diverge from assumed personas and expose the risks of overly simplistic targeting [5:18] [5:32] [6:05].
  • The Say-Do Gap and the Wish-Do Gap: The say-do gap refers to the divergence between what people say they do and what they actually do, while the wish-do gap captures the aspirational self-image that can itself become a brand insight or opportunity [8:35] [9:06] [10:49] [11:10].
  • Data Literate vs. Data Adjacent: Being data adjacent means working alongside data so it informs decisions, whereas being data literate means being able to create research and extract meaning from the results [13:35] [14:58].
  • B2B vs. B2C Research: The underlying principle of understanding customers is the same in both B2B and B2C, but B2B research focuses on people within their professional context rather than their personal lives [16:04] [16:18] [16:58].
  • AI in Market Research: AI is a powerful tool for automating repetitive tasks and accelerating workflows, but it is not a replacement for human expertise, and researchers must find the right balance between synthetic and real data depending on the use case [24:50] [25:21] [25:48] [26:25] [28:48].

 

Transcript

Transcripts are auto-generated.

Kiran Kapur (00:01):
Hello and welcome. This week we are in the world of market research, specifically looking at why data isn't enough and why you need the research to pull things together.

Molly Strawn-Carreño (00:10):
It was in that shop along experience where she grabbed organic strawberries, but non-organic broccoli. And the researcher asked, "Why did you do that?" And she goes, "Oh, well, the organic strawberries are for my kids, but the non-organic broccoli is for my husband."

Kiran Kapur (00:27):
I'm delighted to welcome Molly Strawn-Carreño, who is in Los Angeles. So Molly, you are director of brand growth at aytm. So what does aytm do?

Molly Strawn-Carreño (00:38):
aytm is a market research tech company, and we've since started dabbling into the world of AI, as is a lot of industries in the world. And so we have a really couple of cool, interesting products that are helping market researchers, marketers, anybody who needs to ask questions of their audience, enabling them to do those really interesting studies and gather those actionable insights so that they can make better business decisions.

Kiran Kapur (01:06):
So one of the things that you talk about is that there are things that the best brands have in common and that data just isn't enough. So I'd really like to sort of explore that a bit further. So what do the best brands have in common?

Molly Strawn-Carreño (01:20):
They listen. I think that that is the most defining factor and not just a performative type of listening, not just asking some questions and moving on or supporting a hypothesis, but they treat that research process as an actual way to engage with their clients. So a brand that potentially runs a tracking study once a year and has one ongoing behavioural validation that's built into its decision cycle, that's probably not the best because a lot of brands still say, "Oh, we know our customer and we know this and we know that." Oftentimes they're running off of assumptions or they're running off of very talented, experienced people who perhaps have experience in that particular space and there's not a lot of pressure testing actually in the space. And so the brands that do this really well are in constant dialogue with their customers and they're constantly in the places that they're in.

(02:17):
So if that's a TikTok hack video, if that's a Reddit forum about the new ways that they're using their product or who's using their product or why their products are great or problematic, they're the ones that are going to really have a differentiator in the market.

Kiran Kapur (02:34):
Is that because they hear the real problems or they hear the real rationale behind customer's behaviour?

Molly Strawn-Carreño (02:41):
I think that it's both. I think that a lot of times businesses have a lot of trouble coming to say that something is a problem or that it can be problematic because there's a lot of things that probably went into that release or went into that. And owning that as a business is something that can garner real respect, but it's something that's very challenging and a vulnerable thing for a business to do. And then what was the second part of your question?

Kiran Kapur (03:09):
So is it just understanding the rationale behind the behaviour?

Molly Strawn-Carreño (03:13):
It is in the sense that they do need to understand why people are reacting in the way that they do, but it's a step further than that. It's what is the customer expecting these brands to respond with? How are they going to take that concern, that problem and turn it into something that they actually want to see? And that is where the research can happen. That's where the truly understanding and getting into the minds and the day-to-day of those customers is going to be the most beneficial.

Kiran Kapur (03:47):
Okay. I'm going to ask you about B2B research in a minute, but I want to carry on with the day-to-day customer. So you have a great story that I've heard you use a couple of times about somebody buying non-organic broccoli, but organic strawberries. But it was such a lovely description of why a customer behaves in a way the customer behaves.

Molly Strawn-Carreño (04:06):
So that story was from a guest that I had on the aytm podcast, Curiosity Current, of which I'm a co-host. And she was talking about a qualitative research study where she did a shop along with a mom of a young family. And this mom could say everything about her ideals, the types of products she likes to buy, what she likes to buy for her family. She likes to buy organic produce. She likes to do this, she likes to do that. And from that, there's assumptions to be made. There's types of advertising that could work on her, products that will work on her, things that don't work on her. Maybe she's less price sensitive because she's more intent on buying organic because of the health benefits, not so much the cost. There's a lot of things that can be garnered from just the type of person that she is, the persona that she is.

(04:56):
But it was in that shop-along experience where she grabbed organic strawberries, but non-organic broccoli and went and started moving on with her shopping. And the researcher asked, "Hold on, why did you do that? Why did you grab the non-organic broccoli, but the organic strawberries? That doesn't make sense to me." And she goes, "Oh, well, the organic strawberries are for my kids, but the non-organic broccoli is for my husband because I frankly don't care if he eats organic or not. He's already cooked. He's already the way that he is. But my children, it matters when they're growing up and as their brains are developing that they have this specific type of produce. The kids will only eat the strawberries. They won't touch the broccoli, so it doesn't matter to me." And it was such a human moment where if you were at a mom group or you were asking your friend about that, you would chuckle to yourself and be like, "That totally makes sense.

(05:55):
That's completely rational." But from the research perspective, it was a really interesting divergence from what the researcher expected her behaviour to be. So I feel like as marketers, as researchers, we sometimes pull ourselves away from the actual weird behaviours that people do that make sense as a human being, but maybe don't fit nicely into these persona buckets. And that's where the assumption comes into play where you say, "I'm assuming I'm going to give only organic advertisements or coupons or whatnot to her when in reality she's not just an organic produce buyer. She's an organic produce buyer within the context of her children."

Kiran Kapur (06:38):
That's so interesting. And it's a way we probably wouldn't even think of framing.

Molly Strawn-Carreño (06:44):
No, because when we put on our business hats, something happens to us. We think of ourselves not as people anymore who are talking to other people, but we sort of start to think of ourselves as what message am I delivering to this vertical, which is now it's a thing instead of it's human beings who are nuanced and weird. I always think, who would have ever guessed that in the pandemic that affected us all in tonnes of different ways, that human beings would hoard toilet paper of all things or cans of re-fried beans? I mean, it's things that are so human, but there's absolutely no way to have ever predicted what people's behaviour was going to be like. Sourdough taking off, indoor plants, that obviously became my thing during COVID, but there was really no way of understanding that. And I think a lot of marketers and researchers can lose sight of that sometimes.

Kiran Kapur (07:43):
Yes. And do you remember all those awful. Well, I mean, they were very well-meaning adverts of people staring into the distance and piano music being played in the background and the voiceover going, "We're all in this together," which was true, we were, but actually the ones that worked were things like the Kentucky Fried Chicken adverts where you've all tried making it yourself and you'd have these sort of sad looking things. Because again, it was human. It's what we'd all done.

Molly Strawn-Carreño (08:07):
Yeah. And yeah, the we're all in this together. It's like, must be nice quarantining from your mansion. I don't know what that's like.

Kiran Kapur (08:17):
Yes, there was an element of that. So I mean, the other thing I really liked, you had a phrase about the say-do gap. Now we quite often talk about the say-do-gap. So should we just explore that before I move into the other bit of that comment? Sure, sure. So what's the say-do gap?

Molly Strawn-Carreño (08:35):
The say-do gap is something that's well-trodden territory for anybody who is interested in the behaviour of people, whether that be marketers or researchers or data scientists. And it pretty much means that what people say and what they do is different, and there has to be a sophisticated way of understanding more of what they do because that's a more accurate predictor of how they're going to interact with you or your products or your brand.

Kiran Kapur (09:01):
And then your second half of that was the wish-do gap, which I thought was so interesting.

Molly Strawn-Carreño (09:06):
I think that captures a bit more of the idealistic side of human beings. Human beings, I've found naturally don't like where they are. They have dreams, they have aspirations, they have places that they want to be, and that can be any type of behaviour. Human beings can wish they ate more produce that was healthy for them. They wished they went to the gym more. They wished that. Fill in the blank. Human

Kiran Kapur (09:36):
Beings

Molly Strawn-Carreño (09:36):
Always have a fault that unfortunately we're trying to attain that next goal. And so human beings also enjoy putting that of who they want to be on display for people. And this can translate also too to a research study. So when you ask a question that says, "How many times a week do you go to the gym?" Because people struggle to be honest and they want to put their best foot forward, they want to be seen in a positive light, they're often going to say, so the say-do gap, what they wish the wish-do gap that they did. "I wish I went to the gym three times a week. "And so they might say," I go to the gym three times a week. "But to allow people the opportunity to explain where they wish that they were versus where they are is where you can close that gap even faster.

(10:30):
So a reframe for that exact scenario is how many times do you wish you went to the gym every week? Okay, three times. A follow-up question," How many times do you actually find yourself going to the gym? "And they might say zero, one, for whatever the circumstances. Mine is a big fat zero with a son who's turning one this week. So that can help people put it out there that they have an aspiration of where they want to be, but they have a place that they currently are. And that gap can sometimes be the insight itself. So in this place where maybe let's ask the organic produce question or how many times do you purchase fresh produce versus how many times do you purchase it and then it rots in your fridge, that's also pretty high for me. So asking those questions in a thoughtful way that allows people to express what they wish that they did can also uncover an opportunity for you as a brand to say," Well, this person has an aspiration to do X, and my brand is what's going to help them get to that place that they want to be.

Kiran Kapur (11:42):
"That's so interesting. So you talk about marketers need to be data literate. So we need to understand data, but you also say we have to be data adjacent. And so again, can we just explore that? What do you see the distinctions being?

Molly Strawn-Carreño (12:00):
Yeah, I think it's both, and it can be difficult if you're a creative marketer like me. In the before times before AI, I did copywriting and graphic design, and that's where I started and that's what I studied and everything. I struggled to make the connection between what my work was and the actual impact. If I created something that was beautiful, but it slopped on social media, I would be a wreck. I think most people who get really tied to their work and their personal feelings towards something can be in a difficult place when it's pressure tested against the world. But it's important to not define your worth as a marketer in that way, but learn from that data and learn how to interpret it, understand the dashboards that your RevOps teams are creating because at the end of the day, it's sales and bottom line that is paying your salary and your job is to make, I always like to say, make the ice cream easier to scoop for your salespeople.

(13:02):
This is how I think about B2B tech sales in the world that I live in, which is if a salesperson goes into a room with somebody they don't know, they mention aytm and they say, "Oh my gosh, I've heard of you guys. What do you guys do again?"

(13:17):
My job is done because that means that that little bit of brand perception is going to make their job not banging against ice. It's going to make it easier for them to do. And in order to understand that and my impact, I need to understand the data. And that's just the business data. It does go further than that. So that's what I mean by data adjacent. You understand, you look at the data, you move alongside it, it impacts what you do, you impact the data, but you are not truly literate in the data. I think that that is where we're seeing a shift. We're seeing that market research as a whole, because of AI, because of the benefits that it provides us, you can know absolutely nothing about something and still do something with it in AI. What we're seeing is market research is becoming less of a pillar of a function and it's becoming a centre of excellence, and that centre of excellence can live in many different places, but we're seeing it in a lot of large organisations moving into the marketing function to where now market research, actionable insights, and understanding your consumer are sitting with marketing teams.

(14:28):
I myself just ran the aytm brand tracker this year for the first time, and that is sort of our first little dabble into it because we're like, "We're saying that this is the case, marketers should be doing this." Luckily, I have a background in market research, so it was a pretty easy curve for me, but not for a lot of marketers. A lot of marketers will receive a deck from their research and insights teams and say, "Oh, this is what people think about our products these days, and this is the red package versus the blue package. This was the data for this and this and this and this," and they can take it and make choices with it. But now the creation of that asset, the understanding that the depth that went into the creation of that is now living with marketing teams. And so to be data literate means to be actually able to dive into those numbers, to create the survey, to dive into the data and pull from it what the heck you're going to do with it because it's very easy to see numbers on a spreadsheet that say red versus blue package, but what is the actual reasoning behind that?

(15:34):
What's the choice that you're going to make? Are there more studies now that you need to do? Maybe something surfaced in the red versus blue thing that you didn't even think of, like a certain message type or a font type or something that's very specific that you may not have seen to begin with, and you need to be able to parse that out from the research that you did.

Kiran Kapur (15:54):
You've talked about B2B and most of the examples we've talked about in B2C. So B2B marketing and market research, is it different from B2C? Are the techniques different?

Molly Strawn-Carreño (16:04):
The techniques are different. The bottom line, the meaning behind it is not. So you still need to understand your customers because in B2B marketing is where I've lived for my entire career. I just talk about B2C marketing because that's what our clients do is a lot of the times they are large B2C organisations and do research with us to understand their customer. So it is the same, but it's also a little different. It's the same in the sense that we need to still understand customers' pains, their struggles, but it's just within the context of their business. So where is their business heading? What are their fears in how their jobs are going to be done? We're seeing that a lot in AI in market research and how AI is going to assist the researcher, replace the researcher. I mean, we've all kind of agreed it's not going to replace the researcher, but it is massively changing workflows.

(17:04):
So understanding that, what I was just talking about with understanding the way in which AI is changing pillars of a function, that understanding your customers and how they're changing how they do their job, how they're changing the types of things they look for in vendors and in partners, the types of challenges that they're having with their research, be it a data quality issue, which is huge right now in market research because now AI is making it even easier and more sophisticated to spam surveys and create false data. And so it is the same thing in the sense that you want to understand their problems and their struggles and understand where they're going and be a good partner, but it is different in the sense that it's within the context of their business versus their end goals. So the organic versus the non-organic produce wouldn't really happen in a B2B context.

(18:01):
It would happen in the context of somebody's job.

Kiran Kapur (18:06):
So that's interesting. So you are researching people when we're in our work head space as opposed to researching people being people.

Molly Strawn-Carreño (18:16):
Correct, yes. Yes. Which is fun on its own because you have to understand a person's professional lens, but it's not in their day-to-day behaviour. But things like the wish-do gap still exist in the sense that I wish that I was way better utilising AI in Claude Code than I am. So if somebody says, "How many times do you use ClaudeCode?" Would I fib? I don't know. I probably would say no because knowing my luck, there'd be a quiz that would test me that I would utterly fail, but that would be an example of something. It would just be within their professional context.

Kiran Kapur (18:57):
I suppose what intrigues me about B2B research, I can understand that you can sit with somebody when they're going out shopping and watch what they're actually shopping. Can you do that similar sort of work within the B2B or is it other techniques totally different?

Molly Strawn-Carreño (19:13):
I can chat about the types of research that I did. In my role as the director of brand growth, what I'm most interested in is how people who sit in the market research industry, who are our customers, who are insights professionals, or as I mentioned, marketing data adjacent professionals who are interested in actionable insights, I'm most interested in the way if they have awareness of our brand and the way that they see our brand. So in that sense, it can be pretty similar to a B2C study as a brand tracking study to understand how your brand ranks amongst the competition. So that is probably the strongest comparison point. I don't know, maybe shop-alongs exist for B2B research in the sense of. I know there's a lot of research that goes into how B2B professionals select certain products. So there's a lot of healthcare software and HR software and IT software.

(20:22):
How do these people make choices for their business about what types of products they choose to use? So shop-alongs for that.

Kiran Kapur (20:33):
And so when you are looking at your own brand, how do you go about tracking? You said you're tracking the brand growth and you're tracking the brand awareness. How do you go about doing that?

Molly Strawn-Carreño (20:43):
We've spun up a couple of new different things. The brand tracker is one of them, as I mentioned, and my sort of gold metric for North Star metric for brand is the unaided awareness piece. So if you are asked to list market research brands that you are familiar with is aytm on that list, it can be challenging to see if your brand is or isn't on that list, but from a brand perspective, that's probably the highest metric. The other thing that is, as I mentioned at the beginning of this, was all about listening and understanding what is being said about our brand in rooms that we aren't in. So this can be online forums, this can be in conference workspaces, this can be in all kinds of different places that our customers get to come together, and understanding if we have advocates, if we have detractors, what the problems are, what really the truths are.

(21:46):
And it is a little bit more difficult in the B2B context to really get that because I think my favourite example is the study, the little questionnaire that you would get that says, "How likely are you to recommend Microsoft Word to a friend?"

Kiran Kapur (22:03):
The net promoter score, yes.

Molly Strawn-Carreño (22:06):
I just always chuckle. I said, "Do you really think I'm hanging out with my friends being like, have you tried Outlook?" It's a little bit more challenging in the B2B context to get that kind of data versus, "Hey, have you tried this new lipstick? Have you tried this new hand cream? Have you tried this new fill in the blank?" That's a little bit easier than a work product.

Kiran Kapur (22:34):
One of the things I wanted to ask you was how you ended up in market research, and you've already indicated that this wasn't necessarily where you started your career. So how has your career trajectory moved?

Molly Strawn-Carreño (22:47):
Oh, this is always a fun question because market research is one of those industries that once you learn about it, makes complete sense that it exists, but not a lot of college kids are saying, "I want to work as a market research tech provider." No one's really doing that. I think I've met maybe three or four people in my career that have actually gone to school for market research. Everybody else you talk to found a different path into this space, myself included. So for me, my first job out of college was at a market research company that was a startup at the time, but it was because my alma mater, Cal Lutheran in Thousand Oaks, ran a coworking space called Hub 101 in Westlake Village, and they employed student interns. So by the time I was 19, I was interning there, I was interfacing with a tonne of different founders and learning really cool things and contracting in a bunch of different places.

(23:46):
And one of them was the market research startup where I started working, and I worked there for three years, and then I worked at a different market research firm for three years, and now I've worked with aytm now for three years. And even before I was there, I was doing interning and stuff for the market research firm. So it's been probably 10 years that I've worked in market research and I sort of never expected to be here, but actually I met my husband at that startup, so I can't really complain too much.

Kiran Kapur (24:19):
That's a lovely background. And just finally, you've indicated that AI is changing everything, and I think we all sit there going, "Oh, AI is going to change the world." What's the one really good thing that you use AI for?

Molly Strawn-Carreño (24:33):
I think there's the assumption that you could go to ChatGPT and say, "What percentage of moms buy organic strawberries?" And it would give you an answer. The thing that I've heard with AI is that it is an incredibly good predictability calculator. It can make incredibly sophisticated assumptions, but that has to be fed by real data somehow and some way. I think the best use case for AI as it exists in market research and in marketing is that it's incredibly good at automating repeatable and mundane parts of research. So it can help you get past the blank page to create a first draught of your survey, an initial pass at an analysis, a pattern recognition to say, "A lot of people think this or a lot of people don't think this." But at the end of the day, for marketing and for market researchers who are looking to uncover data and actionable insights, you're selling to human beings at the end of the day.

(25:33):
So in my opinion, you can't use a robot from start to finish to then be received by humans. So content and insights that are created exclusively by machines are going to miss that human touch, that nuance. So in that way, the way that AI exists now in 2026, I don't know what it's going to look like in the future, but in 2026, it exists as a tool and not a replacement for any type of expertise. There's a lot of things we're seeing in market research like digital twins or synthetic data, and that's all fancy ways of saying that you're utilising AI to give you an output versus using a real human being. There's somewhere in the middle that research providers and companies who utilise research are going to have to land. So when you use AI to augment or generate research data, you're making a trade-off between how real the data is versus how much you're sort of protecting this individual privacy, and that trade-off has consequences depending on how far you go on either angle, and so you need to think about where you are on that spectrum of completely synthetic data versus completely using real data.

(26:49):
And there's use cases for each. Let's just say an example is you have 55 to 65 concepts for package design, that's a ridiculous number for a real human being to have to sort through, for you to do rounds and rounds and rounds of iteration to get to your winner. That might be a great place to utilise synthetic data or AI to maybe help you get it to the top 10. And the top 10, then you can go out and test with real human beings. So that's an example of a workflow change. Another workflow change can be you utilise AI to help you create the survey draught. You use some synthetic data for those early testings, you use real human data in another place, you run those transcripts or that raw data into an LLM and it can generate a report for you. Obviously it can hallucinate, it can make stuff up and it can be very convincing when it makes stuff up.

(27:51):
So to go through, use your judgement, read those things, find the sources, challenge it, ask it, "Are you sure about this? Give me the source. Where did you pull from this? Give me all the information and using it as a back and forth tool to then create your final product." So I would say it's not changing the work itself. There's still a purpose that the work fulfils. It's just augmenting a lot of these arduous processes that have been required of research and marketing since the beginning of time. So the speed at which you could do things is faster, but maybe I'm biassed, but there's a lot of people that just have their agentic CMO and they just do some stuff and they connect their social media and they do some stuff. I mean, power to you, but that kind of stuff makes the skin crawl because you can poke holes in so many things.

(28:48):
Whatever you are specialised in, you can poke holes in whatever AI produces for whatever you're specialised in. So it's not there yet, in my opinion. It's really just a tool that's changing the workflows, but the way that the workflows are changing are fundamental.

Kiran Kapur (29:05):
That was an amazing answer. Thank you so much. Molly Strawn-Carreño, who is director of brand growth at aytm, and as Molly indicated, she has her own podcast which she co-hosts, which is The Curiosity Current, which I've been listening to today and be thoroughly enjoying. Molly, thank you so much for your time and your expertise.

Molly Strawn-Carreño (29:24):
Thank you. It was so wonderful to be here.