In this episode, James and Paul unpack the barriers some more traditional marketers may have to adopting technology for market research. Although this technology exists and is used in everyday life, there is still some resistance to making use of AI enabled data collection.
Is this just a dose of healthy scepticism or a lack of understanding of how to use AI-led data to best effect? Listen on to find out how this technology can be a valuable tool for everybody wanting to understand consumer trends without personal bias or opinion.
Listen to the full recording below or on your favorite podcast platform.
Dr James Piecowye: Hi, my name is James Piecowye.
Paul Kelly: I’m Paul Kelly.
James: And welcome to know your audience, the micro mini podcast series. Paul, great to see you again.
Paul: Good to be back. Thanks, James, have you been?
James: I’ve been unpacking everything we’ve been talking about. And now I’m constantly looking at grocery stores, I’m looking at retailers, and I’m trying to figure out what kind of research they’ve done and trying to figure out if they’ve gotten it right.
Paul: Figure out who’s gaming you.
James: Exactly! And with that whole mindset in play, it just got me thinking that when we had an opportunity sit back down, as we’re doing now, to start thinking about this whole universe of research that where we’ve opened up the door to and what gets in the way of every organization starting to adopt some level in some form of AI enabled data collection.
Paul: Well, I think it’s an interesting point, because I think there’s a quite a lot of reasons, but a lot of them are often just organizational and the lack of, I guess, embracing change, probably well founded, you might say, skepticism towards towards some versions of AI because they’ve been, for instance, a decision maker has been to a conference and heard someone talk and it sounded complex sounded far fetched, to some degree. I think, as we’ve touched on in previous episodes, we’ve always or sorry, in our day to day life, not always recently, have embraced AI without necessarily knowing it by using our smartphones by using predictive texts by using Gmail, if you use Gmail, you know, that feature of quite well, driving a car, for instance, if you’re using almost all modern cars now have some sort of collision detection system, that type of thing. Even parking sensors to some degree are smart.
James: So is it just trust Paul?
Paul: It is and what people don’t realize is all that stuff that you use every day like programming, a smart, smartwatch or whatever. It’s the same technology, it’s just being used for a different purpose. And I think trust comes into it in the sense of, we’ve always done things a certain way. We’ve always gone and interviewed people.
James: Well, that so let’s back up for a second when we think about doing research, and I’ve got a product, we’ve talked about different products throughout this series. And I want to now find out why are people averse to my product. Why are they buying the other guys? Or what do people really want from this thing? You really nailed it here in that we’ve got a stable set of tools. And our guests have spoken about this that we’ve been using for 100 years. Do you think it comes down to just comfort level?
Paul: Yeah, it does. Yeah. And I guess, you know, it’s good to be skeptical, I guess of new technology and don’t necessarily want to be an early adopter for it not to necessarily work out for you. And that’s fine, I think. But we’re sort of beyond that stage. Now. I think we’re well beyond early adopter anyway. And we’re probably still in a, you know, very nascent phase of, of utilizing technology for understanding people better. But yeah, it’s, it’s, it’s definitely a case, if that’s how I’ve always done it resistance to change. It’s too much to understand. I think for a lot of people, I’m not being facetious. I’m not being dismissive. Sometimes things are made very technical, that don’t need to be made very technical.
And if you’re able to see the underlying data, where, where things where conclusions have been drawn, you might have various objections, for instance, for doing this sort of research, and I almost, by the way, guarantee that finance departments in the very same company using AI to go over daily transactions. They’re not, there’s nobody there anymore sitting coding manual transactions in a ledger anymore. Machines do that. For us. It’s a given now, it’s not even questioned that there’s AI technology helping us with accounting, for instance, but in the consumer insights, or research or product understanding or marketing functions, it’s still a bit of healthy skepticism. But we’re moving beyond the realm I think of it being an early thing and being able to embrace the change.
But understand the technology is a leap. And it’s not a leap that people necessarily need to make, you don’t need to know how the code comes up with it. You just need to understand, I guess, on a basic level, how it works, and whether it’s right for you, and to be able to see the underlying data as well. And that’s, I mean, those three things should counter most objections because I think most market research people, at least that we talked to have a healthy skepticism of how things have been done forever. So they’re more like, well, we know people lie when not intentionally, but we know people lie when we ask them questions. We know they don’t behave, how they do every day when they when we visit their houses and that type of thing. We know that, but there’s nothing else. And it’s like well, there is technologically advanced way that we can do this?
James: In one sense. I’m going back to what Faisal said in a previous episode. There’s a healthy medium of new technologies and old, tried and true techniques being used. I want to stay in this vein for a second, I know you’re saying that it’s that this technology is being adopted by companies. It’s being used in a variety of different departments. We’re using it every day, and we don’t realize it. I wonder when we talk about the uptake of this technology and the use of AI for audience sentiment research, do you think in a sense that it’s a young person’s game at this point?
Paul: Yeah, no, I think so. I think it’s a complex question, because there’s organizational barriers, there’s also personal barriers. And then there’s also previous disappointment. So an example of what I mean by that would be something like a social listening platform where you have to enter the queries yourself. So you have to enter the Boolean style operators, basically, you got to know how to do that to get it right to begin with. So first of all, you got to have an understanding of I guess, of advanced queering like, and all those types of functions, as well as understanding that you may have inherent biases in the keywords that you select to find information, I would suggest, respectfully to most people who would listen to this that probably six out of 10 cases don’t work out with social listening, because of that.
James: Six out of 10?
Paul: Yeah, and it’s qualitative. There’s no quantitative element to that. But there is sort of the homework hasn’t been done, it’s complicated. Like you need to understand how a query operator works. And you also need to understand how to spot your own biases, and that you’re not looking for, you’re not using it for confirmation bias, you know, I think, because my friends, use, I don’t know, eggs on their face that, that that’s the trend, and I should be, you know, I should be recommending that to my boss, that we start marketing eggs as a face treatment, rather than a food or something, you know, obviously an absurd example, but I’m just saying, it could be three, literally three people who have that conversation in your circle of friends, and you suddenly think, Oh, that’s a trend. And you’ll start to look for confirmation of that. That’s the risk of those sorts of things. And unless you have some training to at least spot that, and also the training to construct these queries properly, then you’ll be a bit disappointed.
And so I think a lot of the time there’s a skepticism towards, for instance, using social as a data source because of that reason, because I because I don’t know how to use it. It’s not working out properly for me. And that’s reflective then of all solutions within this space. That’s what happens. So it’s not necessarily potentially a youth thing. Or at least being a digital native versus a digital adopter, versus somebody who started their work before the digital era, for example, like, I don’t think it’s as much that because I think a lot of people, particularly older people understand that technology enables things that just weren’t possible before, whereas younger people potentially take it a lot more for granted. So I think I don’t think that’s like an enormous barrier. But I think previous disappointment, perhaps with various tools or platforms, does become a barrier, and introduces some skepticism, which I say, is healthy, I think it’s healthy to be skeptical about where the information that you’re getting to make decisions comes from and the validity of that. But you also have to trust it when you see it.
James: How do you talk to people about the fact that by using these tools, you are opening up the potential to be examining an extraordinary number of data points and getting a huge amount of information that other forms of research that one might have in their tool chest: surveys, interviews, observation, much fewer data points, how do you talk to people about understanding that you’re going to have to do some work to wring that very large field into a manageable plane?
Paul: I think, first of all, that the numbers are beyond sort of some comprehension. And what I mean by that would be that, you know, if you’re talking about millions of people, it doesn’t make. it’s hard to comprehend. It’s a bit like saying, you know, how much does a million dollars weigh? You know, there might be 1% of the population? Who’s felt that in their hands before, but nobody else has, right? Particularly, I guess, these days with everything. Numbers are on the screen. But what I mean by that is that I think there’s a large proportion of people who sort of get to that number and go, Wow, that’s impressive. But, you know, how’s it any different to 2000? And that’s when you start to say, well, we have to start to look at the data.
We have to we have to transform that data. We have to understand it, but what it gives us is a bigger cross section. So you’re taking a a representative sample of say 100 People with our representative sample as a couple of 100,000 we can be, we can have a greater degree of confidence in the outcomes. There’s not really false positives, because we’re not sort of dealing with people who answer questions on a professional basis, we’re not dealing with people who are acting differently in front of us. In fact, we’re observing people over a long period of time. So I think we’ve touched on historical data or trended data, as it’s called in market research. And that gives you a greater confidence about the validity, I guess, of the numbers and the size of the audience. But it can be difficult to comprehend a bit like huge numbers that we might read about federal budgets as well, you know, like, the US federal budget is 7 trillion or whatever it like. What is that? What is that? Yeah, yeah. And that, that’s a very similar scenario. I think when you sort of say to somebody, well, look, well, we’ve looked at 240,000 People in this survey, it’s like, well, you can’t mentally they’ve been that’s three large sized stadiums, right? It becomes sort of a little bit impossible to comprehend. But then it gives you that extra degree of confidence about the outcomes and the recommendations and the actual trends that you’re saying, legitimately there. And that’s a huge step.
James: Look, I know, in your own in your own organization, you’ve had clients come and go, is there a point with this type of research? Is there a group who might not actually need it?
Paul: There is a space, I think, for everybody, it’s very different set of circumstances that somebody might use it across different industries. But if you sell anything to anybody, whether that’s another business, or it’s a person, or well, it’s only gonna be one of those two things, you if you do those two things, then you can benefit from understanding the customer a lot better. And that’s a number of things, you need to know who your customer is to sell to them. So that’s, I guess, the baseline of selling, selling 101. But also, it’s about honing your products, understanding if your product has a fit the market, is there a market there? What do people think of your brand. Brand, obviously equals sales, there’s lots of things where that’s beneficial. And I think that’s across the spectrum from somebody setting up in their home making cakes or selling and selling on Instagram, through to through to multimillion dollar, multibillion dollar sorry, international companies, there is a space for it.
That being said, what you spend on it, obviously has a very different set of outcomes. And, you know, if you don’t have the budget doesn’t make a lot of sense, you should be using the baseline of marketing the core marketing principles to reach as many people as you can. So people can think about your brand and understand your brand and understand your product and just go from there. So I guess, if you’re talking about people starting out, it probably doesn’t make a lot of sense. But it might also make a lot of sense to do some trend watching the number of good platforms available, I think we’ve touched on it before that might be 50 bucks a month or something like that, you could do that for a couple of months and get some great results and understand what some trends are, or some emerging trends that could fit to your market.
Also, when you sell directly to people, you have a better idea of what they actually want. So but when you’re talking about shifting cans of soft drink across hundreds of 1000s of retailers in the 10s of millions in quantity, then you don’t have that ability as much. You get feedback, but you know, it’s not the same. And then, yeah, I think in terms of who’s right for and who’s not right for, it depends on the organization or embrace of that. To be honest, James, so people have to be embracing technology has to be a technology forward organization, because it just doesn’t, it just doesn’t gel, if you’re I think we can all picture perhaps I’m gonna use a stereotypical example of perhaps a bank or engineering company or something like that, that might be a bit more traditional, less digital, the most digital they are is perhaps emails and things like that. And we know, there’s, you know, everybody tries to transform and everything like that, but just some organizations don’t have that technology mining.
And if you don’t, you’re going to be probably skeptical of the data to begin with. And that’s going to cause problems when you try and adopt it because you’re always going to be second guessing the data and what’s in front of you, and you’ll go with your own biases or in opinions. And so I think it doesn’t work in that circumstance. But by and large, it’s for everybody. It’s just different scales require different things. And if you’re starting on your journey, then most definitely I would recommend looking at trends and things like that, but perhaps, you know, a small software as a service type platform that’s not going to cost you the earth just to aggregate trends on Instagram or Tiktok, or whatever, get an idea of who, who’s buying what, where, and close your account and use that knowledge to help hone your messages. And then I guess at the other end of the spectrum, it’s about if you’re, if you really embrace technology, and you’re already skeptical of your consumer understanding, then embrace the technology. And that will that will change everything,
James: I’m just amazed at how fast this area’s evolving. And, you know, as we’ve talked about, just with the the Arabic language, and going across dialects, and how quickly the technology is learning, and the ability to give me insight that I can make better decisions with, and as you said, whether it’s a mom and pop in the basement, or a multibillion dollar corporation, the entry levels and the ability to scale is so easy and affordable.
Paul: Exactly. And I think the analogy before when you know, we were discussing an earlier earlier point about accounting software, it’s the same thing, so that some of that stuff is built for mom and pops in the basement, you know, to keep the keep track of their sales tax or keep track of the loan amount amount of transactions that they have. And, you know, that’s nine or 10 bucks a month or something like that. And that world was previously unattainable. Not even that long ago, 10 years ago, there wasn’t really the software that could just connect up to your bank account carried all your transactions, and, you know, most most like imagine saying to a mom and pop business maybe even five years ago, look, you can take mobile payments from your phone, from a car payments, you know, it’s the stuff has just changed immensely. And that adoption of that technology is sort of without question. And then and then on the other side of the fence, I guess it’s a little slower. And that’s mostly because, as we’ve discussed in previous episodes, it’s been done this way, for generations now. So so there’s a there’s a shift that has to happen. And that shift is very much linked to how technological companies,
James: What I find interesting is looking at, at different companies and we’ve talked about this in the past. And in season one, actually, you had a great example of a company who was had issues with a product and you ran that an AI model on it, and said, Here’s the problem. And they said, No, I can’t be. And so then they went, they went back and validated your suggestions through literally doing in person observation, and then came back and said, You were right. And I think to me, that that becomes kind of the glue in that the AI tools can work very well with the traditional tools to add some validation and actually open up the insight, because, as you said, their data was there, their traditional data was all over the place. And they couldn’t account for why people weren’t buying this product that they should have been buying.
Paul: Yeah, yeah. And it was, I think, in that in that particular circumstance, it informed the brief of the traditional research and inform the brief. it reframed the brief of marketing, then to go back and think about the messaging and how, you know, certain things were being talked about how education was, you know, just just opens up and tightens briefs and for the the outlay, and the costs, the future sales returns become immeasurable against that outlay, if that makes sense. So the small outlay cost for the Insight has outweighed return in sales, and that just once once, once that’s seen, then widespread adoption of technology becomes obviously much easier. There’s no barriers, but also, as you say, it just informs more traditional methods to make them tighter, to know where to start looking, and then to make sure that the answers can be validated. And if they are validated, then it’s great. And if they’re not, then what’s wrong and who’s right. And it’s a lot easier to work out that way, you know, rather than just hearsay, you’ve got the data.
James: What I find interesting is that if we’re talking about the sentiment of a user of something, yeah, I can sit and I can watch them. That’s usually expensive on my time to then transcribe my my observations, put it all together, present it to you. Whereas we’ve got this great tool that can look at what’s going on across thousands of data points via our socials to see well what are people liking what are people liking? Why are they liking it? Why aren’t they liking it, looking at the images looking at the emojis? And in the case of the Arabic world playing with the dialects, so suddenly, it makes me giddy.
Paul: And you’re just opening a bigger world and I think that’s the most important In part of, I guess what was what we’re saying is, you get access to information, everyone has access to information, what you do with that information, and what’s next with that information is the most critical part. We’ve talked about that multiple times that data is literally information. It’s not, it doesn’t it’s not an insight until somebody looks and understands and finds the why as to a certain behavior or a certain attitude, or certain whatever it is the why becomes critically important. And I think that’s what technology enables us to do at a much faster speed and a much bigger scale and a much more interesting way and enables us to know, well, like the title of the podcast “Know the audience”. And once you know your audience, no matter what you’re doing, then you become successful.
James: I look forward to our future episodes where we’re going to sit down and have conversations with people who are using these tools and talk about the use of the tools in the context of what they’re doing. And I’m really interested in what made them jump in. What what they’re getting out of it. Because to me, the biggest challenge becomes I’ve now got all this data, do I have information overload? And how do I know where to scale? In that, that data? And I think talking to people who are doing it opens the door to others who are, who are skeptical, who are on the fence, or who are trying to imagine how this applies to them.
Paul: Yeah, it’s very exciting. And I think, yeah, there’s an unlimited potential. And sometimes that’s a barrier in the sense that the potential is infinite. So you have to be very focused. But once you once you sort of, once you once you know what you want, or what know what you need, or know what your problem is, then obviously that becomes a lot easier to solve.
James: Paul, this has been a lot of fun. And I think we’ve, we’ve nicely brought together our two or two seasons, in a sense, with a nice pivot to where we’re gonna go in future episodes. So I look forward to sitting down and doing more of these. Thank you very much.
Paul: Thanks, James. Really appreciate it.
James: I’m James Piecowye.
Paul: I’m Paul Kelly.
James: And this is the “Know your Audience” Micro Mini podcast.