Market Segmentation: Target Markets vs. Databases (Full Transcript)

[00:00:05] Speaker 1: Hey friends, Catherine here, and today I'd like to talk about market segmentation. You know, for a long time, market segmentation was something that, well, was really in the domain of professional market research, and so those of us who are in the market research and insight space for many years, for decades, we kind of owned market segmentation as a phrase and as a methodology. However, things have changed. These days, when you read articles or even see YouTube videos about market segmentation, it can be about something that's completely different than what we in professional market research think of. So here's some things we need to keep in mind about how the language is used these days around market segmentation so that, frankly, we understand what people are talking about, but also so that we can make sure that we're not going to inadvertently confuse clients or colleagues in the way we use the phrase market segmentation. So first of all, I have a question. When did the definition of market segmentation change, right? For a long time, it really was a very clear definition. So what I'd like to do is let's talk about what we think of in professional market research, in professional customer insights work, as the correct definition of market segmentation, and then we're going to talk about some of the things have shifted so that we can understand how other people are using the phrase. So some of these other people aren't necessarily in our field. They might be in other fields, but they're marketing related or sales related. So these are areas where we will have intersections, and again, it's part of the confusion that is out there. So when did the definition change? Well, part of the question about when the definition changed is about when new sources of data became available. So when we look at how different people use the phrase market segmentation, what we're going to see is that as new data sources became available, those data sources were used to segment target markets in different ways than professional researchers have conventionally done. So really, it's something that's changed really dramatically in the last two to three years. So I would say for myself, up until about two or three years ago, if I use the phrase market segmentation in a professional context, we were all usually using the same meaning. But over the last couple of years, it really has changed, again, because of new data sources. Let's start off with a definition. When we talk about market segmentation in the context of professional research, what does it really mean? Well, at the highest level, the goal of market segmentation is to take a large, broad target market and divide it into logical buckets. The idea is that rather than treat a large target market as a homogeneous group, we should figure out what the important variations are within that larger group. Are there differences in attitudes, and brand perceptions, and product behaviors, and shopping behaviors, other sorts of attitudinal and behavioral information that will allow us to break a large market into logical subgroups. Because the truth is, if you are targeting a large customer group, it probably isn't homogeneous. There probably are important differences in attitudes and behaviors. The idea is that we want to identify important variations in attitudes and behaviors so that a company can tailor its products, maybe its pricing strategies, or how it promotes its products for maximum alignment. Rather than send the same product with the same pricing and the same messages out to the whole world, let's tailor them so that we are sending the right things to the right people in a way that's going to resonate with them. That's the idea. At the highest level, the goal is about sales. This is about making money. To maximize sales to large populations of customers, it's usually best to divide that group into logical subgroups so that we can send the right messages to the right groups. Maybe one group is more price sensitive. Maybe another group is more about high-end features. There are a lot of different ways that we can segment a target market. That seems like a pretty clean definition. For a long time, that's really how those of us in professional market research have defined market segmentation. Well, frankly, it's still how I define market segmentation. But again, things have changed over the last couple of years. So we need to be aware that when we hear people say market segmentation, this is not necessarily what they're thinking about. They're not necessarily thinking about a broad target market and using research to break it into logical subgroups for the purposes of targeting or refining products for specific customer groups. So there are two elements that we have to keep in mind where there is some confusion these days. First of all, when you hear people use the word segmentation these days, there can be differences in terms of what they are segmenting. In market research, we typically think of segmentation as segmenting a broad target market. So maybe I'm doing a study and I am segmenting the market for bottled water. And so my target market might be people who buy bottled water. That might be my broad market. And I might be looking to divide that into subsegments so that I can promote my bottled water products to different subgroups very, very efficiently. So I might be segmenting a target market. However, these days, very often when you talk to people about segmentation, they're talking about segmenting a customer database. Now, if they have a large customer database, that can be a huge population. However, it is a specific population. It's already their customers. It's not potential customers. So in market research, very often when we think about segmentation and segmenting a target market, we assume that that market may include current customers, lost customers, potential future customers. But if I'm segmenting a customer database, then I'm making a decision that I'm segmenting our known customers, people who already know us, people who have bought from us. It's a particular population. It's not a bad population, but it's not the whole potential market for your company's products or services. However, again, over the last few years, a lot of companies have done really cool work segmenting their customer databases. There's been a lot of improvement in how companies segment their customer databases over the last few years. And therefore, when you hear people talk about segmentation, it's a good idea to clarify, are they segmenting the overall target market or just their database? And here's the one that's really interesting. If you go onto YouTube or other social media sites and enter the phrase market segmentation, you'll get a lot of content that was created in the last couple of years about segmenting email lists, literally just the email list. So sure, a lot of organizations have big email lists. That's fantastic information. But again, it's not your total target market. If they're already on your email list, they've already opted in. They're already aware, have some level of favorability towards your brand and so forth. So it's a really interesting list, but it is not your whole target market. So a lot of the times when you see references to segmentation these days, people are talking about segmenting an email list. And we just have to be careful that we understand that these are the three different things people could be segmenting. So if I'm having a conversation with somebody, especially somebody who's outside of the professional market research and insights profession, I need to clarify. They use the phrase segmentation. What do they mean? Are they segmenting a target market, a customer database, or an email list? All three things are very common these days. Then I need to understand what they really mean by segmentation. What is the specific task that they're doing? Are they creating the segmentation model? That is, are they deciding what variables are going to, or what rules, if you will, are going to be the basis of putting people from a large bucket into smaller buckets? So if it's a customer database, it could very well be about how much purchase behavior or purchase activity that people have engaged with. They may divide their customer database by whether or not these are people who are buying frequently or infrequently. They may also be dividing the customer database by demographic variables that they have in the customer database. So depending on the customer database, they have some really cool variables they can work with. But if you're creating a segmentation model, then you have to really think carefully about what that source is, right? But you still just, you know, a word to the wise here. When people say they're doing segmentation, it doesn't necessarily mean that they're creating a model. Often these days, when I hear people use the term, oh, we're doing a segmentation study, what they're actually doing is profiling a given segment or a few different segments. So they've already decided what the segments are, and now they're doing a segmentation study that those of us in professional research would think of as more of a profiling exercise. So the segments have already been defined, maybe based just on assumptions or an a priori model, but the segments have been defined, and now the, quote, segmentation project is really just about profiling the segments. So all of these different exercises have merit and are applicable for different things. However, we in market research need to be aware because when people talk about segmentation, it really can mean different things. So clarify, what are they segmenting? Is it a target market, a customer database, or an email list? And what's the task? Is it really creating a segmentation model or profiling given segments? These are really different tasks. So just things for us to be aware of. Now here's a question. If you are doing a, what I would consider a true market segmentation study, I'm trying to figure out the best way to divide a large target market into subgroups for effective marketing purposes. What are my best sources for that kind of project? It's not necessarily survey data. So these days when I'm doing a true segmentation study, when those of us who do professional research do a true segmentation study where we're actually creating the segmentation model in addition to profiling the segments, then very often we have a choice. We can be working with survey data or we can be working with customer databases. So if I'm working on a segmentation study and I'm using survey data, well, this is great because I'm getting a chance to segment not just current customers. I'm getting to segment the total potential available market. And that can be great. And so survey data is a great way to collect information about behaviors and attitudes that can often be used for creating really rich segmentation models that really reflect the needs of the different target markets, right, the different subgroups. So needs-based segmentation can be a really cool strategic thing for a brand to invest in. However, there are definitely cases where segmentation can be done using a customer database, especially if the goal is to assign people in the customer database to segments for tactical campaigns. So sometimes I'm talking to organizations that are really interested in doing segmentation, but they're doing it because then when the model is done, they want to make sure that all of the customers in their database are assigned to the right segment so that specific campaigns and activities can be initiated to those specific people. Those specific email addresses, those specific phone numbers, those specific postal addresses, they want the people in the database to all be tagged to a specific segment for tactical marketing purposes. It's a completely legitimate thing, and we see it very, very often. The question becomes, which data source do you think has better quality? If I'm trying to understand a broad market so that I can divide it into logical segments, again, so that I can maximize sales by having finely tailored messages that are going to resonate with segment A, finely tailored messages and product offerings for segment B, finely tailored messages and products for segment C, etc., what data source is going to be best? Well, the truth of the matter is that they both have merit. Survey data is great because it gives us attitudinal data, which most customer databases frankly don't have. And attitudinal data is where we get that why information. You know, what's important to people? Why are they behaving in certain ways? So attitudinal information typically is going to come from survey data. Survey data, of course, can also collect behavioral data. But getting people to self-report certain types of behaviors is difficult. You know, when we ask people about recent past purchases, sometimes their recall is not perfect, either about what they've bought or how much or, and I hate when I see this in surveys, what they actually paid. You know, getting people to self-report what they paid, whether they used a coupon, that's really, we know that that's always been a bit challenging in terms of data quality, right? However, a customer database has that stuff. So organizations that have really rich customer databases, especially those that sell a lot through e-commerce so that they've got a lot of behavioral data, that's not self-reported behavior. It's not subject to the bias of self-reported behavior. These customer databases often have very rich actual, even historical purchase behavior. We understand seasonal purchasing, you know, time of day purchasing, what they're purchasing, the price paid, whether or not they used a coupon, et cetera. So behavioral data from customer databases is often really high quality, but the customer database, what's it lacking? Typically the customer database doesn't have those attitudes, you know, we don't, we can't ask them about their perceptions of brand A versus brand B, which can sometimes be important. We can't ask them why they use a certain type of product versus another type of product. We can't ask them about whether or not certain things are important when they're buying certain product categories. So a customer database has often got great profiling and behavioral data, but we often need survey data to get at attitudes, perceptions, perceptions about brands, et cetera. So these days we do see a lot of cases where the segmentation project is using a combination of survey data and customer database, right? So that we have two data sources and we can integrate them for the purposes of creating the segmentation model. So this is something a lot of organizations have been doing for a few years now. It's become, well, I think at the first time I saw it was honestly probably about 10 years ago, but that was back then it was still fairly leading edge. But over the last few years, it's become way more common to use a combination of survey data and customer data for the purposes of creating a segmentation model. So it really is common to see an integrated data approach for market segmentation. So the phrase market segmentation has evolved over the years and depending on who you're talking to, it is going to mean different things to different people. So still a really super important application in the world of marketing. A lot of organizations invest in market segmentation. I personally find it to be a fascinating area of market research and I've done a lot of market segmentation studies in my career, both on the consumer side and the B2B side. And it really can bring enormous strategic value to organizations. It's just that the way the words get used these days, unfortunately can cause some confusion. So when in doubt, if you're talking to new colleagues or clients who are talking about doing market segmentation, just remember, make sure you know what they're trying to segment. Is it a target market, an existing database, or an email list? And when they say segmentation, what are they doing? Do they want to create a segmentation model so that they know what the rules are for assigning people to different segments? Or are they just taking an existing way of defining segments and now their project is about profiling? Both are legitimate tasks, but again, it can get confusing. I hope that conversation was useful for everybody. And for those of you who are planning on taking any courses at Research Rockstar over the next few months, I want to let you know that there's a course coming up in July that I'm actually personally teaching this course. I know some of you have been taking courses with some of our other instructors, but this one I'm teaching myself. It's called Data Visualization 101. And it's going to start on July 5th. It meets on July 5th, 12th, and 19th from 11 a.m. to 12.30 p.m. Eastern. Data Visualization 101 is a great course for people who are creating market research reports and you want to break the bar chart habit. If you find that your market research reports are just got too many boring bar charts, then it's time to expand your library of data visualizations. And so in this course, we have a lot of fun, we get hands-on, and you'll learn about lots of different ways of visually displaying data that are perhaps a little bit more current or might help to more ignite your client's interests, including different data visualizations for specific types of data that you're trying to convey. So whether it's just something simple like conveying parts of a whole or doing data visualizations to show the relationships between two or more variables. So if you find that your reports are just very heavy on the bar charts, this will be a fun, fast-paced class that will really help you to amp up your research reports. If you have any questions, if you have any comments, please do leave them here. And as always, I really appreciate when people subscribe or give me feedback, so please don't be shy. And if you are looking for any other information, please do visit us at training.researchrockstar.com. Thanks everybody and have a great day.