 Hi everyone, we're back. This is Dave Vellante with Jeff Kelly. This is theCUBE. We're wrapping up two days of the Tableau Customer Conference. We've been going wall to wall yesterday and today. This is a great event. Tableau's all about using visualization to help customers gain insight to data and really is trying to change the way in which data is perceived, the way it's used. Essentially, Tableau's on two attack vectors, I would say. One is going after the sort of traditional BI every time Christian Shabo talks about traditional BI. He slows down his cadence and it's a great little marketing tactic that he used. So that's one because we all know traditional BI is like a snake swallowing a basketball. It's just very slow. And then the second vector is, of course, everybody's using Excel. It's great, but it's not. We all have the love, hate with Excel. So Tableau's really trying to fuse that gap between data and insights and doing so with visualization. So we've been covering that all day. We've been talking to customers. We've been talking to partners. We've been talking to executives. Susan Byer is here. She is essentially audience audit. I mean, you are the principal, the founder, the staff, the whole deal. Yeah, there's actually two of us, but yeah. I'm half of the company. So welcome to the queue. Thank you very much. So audience audit, tell us about audience audit. So we're a small market research company. We only do attitudinal segmentation research. That's all we do. And the company, I started it because I spent 25 years doing marketing strategy. And I saw whether it was agency side or client side. I saw lots of marketers making decisions without a lot of information to back them up because their clients didn't have them. They didn't have them. And so we marketers tend to think that we can do things with a good gut. And some of the time we can, but there are a lot of decisions that are not based on anything which drives me personally crazy. So I started this company to help marketers get some better information for marketing strategy. And so that's all we do. We do work for marketers. So you mentioned attitudinal segmentation research. And on your website you have a little pie chart, exploding pie, what the bleep is segmentation. What the bleep is attitudinal segmentation research. So lots of folks now become familiar because of the whole big data thing with segmenting customers. But a lot of them are still doing it based on traditional measures like demographics. So how old are people? How much money are they making? What zip code do they live in? Or now they're using behavioral data. So where are they clicking? How long are they spending on site? What email are they opening? That kind of thing. Our work is different because it segments customers based on what matters to them. What's motivating them? How they feel about themselves and their category? Their pain points? And those kinds of things. And so we segment them based on that. And then we look at all the other stuff to see where there are correlations. But I'll tell you more often than not, we see comparable demographics across attitudinal segments. And the reason is that people don't make purchase decisions based on what they look like. Or how old they are or how much money they make. You can line up a bunch of people that all look exactly the same on paper and they'll all want to go someplace different for vacation. And it's because what they want out of vacation is different, right? Some like a big city, some want to go to the beach, some want to read, some want to go to galleries. And you can't see any of that on paper. You have to understand that by talking to them. And that's the information that marketers need. So a lot of us as marketers, if we have data at all, it's demographic data, behavioral data, it's what people are doing or what they look like. But in order to do things like build personas or drive marketing strategy or figure out effective messaging, you have to know what they care about. And if you don't have that data, you have to guess. So as marketers, we end up making these assumptive leaps about what's motivating people, which is where I believe you get strategies along the lines of, well, they're women, they have young children and they work. So they must be stressed, tired and motivated by an interest in safety. And so you make these leaps about what must be motivating people based on what they look like. And in fact, some of those women may not be motivated by safety at all. Some of those women may not be stressed and tired, but we're sort of lumping them all together because all we know about them are things that lump them together and it has nothing to do with what's actually motivating them to be interested in your brand or your category or anything. So that's great, I love that explanation. Let's go to an example. So let's take one that we can all relate to, which is travel, right? We all travel, vacations or whatever. So you're saying the traditional way to do it would be, okay, give me the age demographic, disposable income, et cetera. How much money do you make? How much vacation time do you get a year? Where do you live? You would say, no, no, no, no, no. Let's look at some different factors. Like for example, what? Would you rather go to museums or would you rather hit the beach? Absolutely. Are you interested in, I did a study like this for a client that was a resort destination. And we saw people who wanted to go there for the outdoor activities, horseback riding, hiking, all of that kind of stuff. And for those people, the availability of those activities and the weather were the most important things they cared about when they were trying to pick someplace. What was the weather like? Cause they're outside. There's another group for that same destination that loves going there and never wants to be outside because they're in the spa or they're in the galleries or they're going out to dinner. So they still love the same place but they want radically different things as part of their vacation. Some people are taking their family with them and they want a vacation that's family oriented. Some of them have a family. So if you ask them on paper, yes, they have three kids but they are not taking them on vacation with them. So understanding. Margaritas in the beach and the up till 2 a.m. Party, party, party. And the problem is, if you don't understand that difference you're speaking to all of those people in exactly the same way. So you. And you're saying it's sunny here and we have galleries. And the problem is you're missing the mark for both of those audiences because you're talking about stuff that's not relevant to them. And so they think maybe you're not relevant to them and then you lose them. So your philosophy then would be to take the conventional way in which you might organize segment in audience. Throw it away. Yes. Start over. So you auto, essentially auto segment for every problem or every opportunity. Yeah, I mean every study is different but we find out the things attitudinally that clump people or organizations together when they're making a decision about a purchase and differentiate them from other people. And a lot of times this work used to be done with things like focus groups where you'd get small groups of people together. I've been behind the glass many, many times. And you talk to them and that's fine but it's really scary to risk the strategy for the next three years for your $35 million business on eight people in a conference room. So we do this work. We do this work quantitatively now. Come on. If you're going to use focus groups properly you wouldn't do that, right? Well, you would be surprised. Well, you should use it to figure out what questions you want to ask but there's maybe better ways to do that. I wish you were right. But unfortunately a lot of people are making decisions that way because they don't know that you can get attitudinal information quantitatively, right? They don't know that you can get to these questions and these answers with statistically reliable numbers that tell you what the attitudes are out there that you should be worrying about and which ones you don't need to worry about because they're not affecting a large enough percentage of the group you're trying to reach. So is it as simple as asking the question straightforward what is it that you're looking for in this vacation or whatever the case may be and then saying, okay, here's the answer. Here, marketer, use this to target these people or is there more to it in terms of analysis? Yeah, if it was that simple I wouldn't have a business. Right, well today of course Apple is announcing their iPhone and Steve Jobs said I don't do focus groups. I don't care what they say. People don't know what they want. I know what they want, I'll create it. Right. So how do you actually, so if you're asking people what they want, Steve Jobs would say, well, that's not really that important. Because I'm asking them about a lot of different things and how they feel about them and the mathematics that we're using are telling us which of those things in combination are important to which groups of people and how big those groups are and whether you should worry about them. And you as an individual wouldn't necessarily be able to identify for me all of the core things that go into your vacation choice. If I sat down next to you. But if I ask you about a huge range of things about how you feel about going on vacation, why do you go on vacation? You know, I can parse out what's really affecting it. And when I reflect that back to you, you'll go, oh yeah, oh that's right. But you can't necessarily come up with that on your own. So it's a little more complex. Right, you're correlating all the information to come to a larger premise or a larger idea. Not just straightforward, here's what they say they want. And honestly, a lot of it is about just understanding the kinds of things that affect people's decision making that they may not recognize as affecting their decision making, right? Now what's your methodology? So we talked off camera. You got to get a good list, right? You can't just talk to anybody. And if a client has a good list, great. You use that. If not, you have ways to find them. That's cool. Should you do this on the phone? No, online surveys. And there are reasons to do things one way or the other. Our surveys are fairly long. Our surveys are on about 15 minutes, which is long for a survey. We have good reasons for doing that. We get very good response rates when we do it. But you can't do that on, you can't ask people to rate 60 items from one to seven on a phone. They have to, you have to be able to see it and you have to rank it up in little chunks. You have to visualize it, rank it, and then you've got little tools to skip patterns and all kinds of cool stuff. Absolutely. And your surveys are fun to take, or how do you get people to stay on you and send them? Well, sure they're fun to take. You know, it's always surprising to me, people are stunned that folks want to participate in survey work. I see people all the time who like being heard. They like having their opinion personally. I do, I take surveys all the time. I mean, when I don't have time, take the survey, no, sorry, I don't have time, but I have time, I'd love to do it. Right, now we're making sure that we're talking to people who actually care about the subject we're asking about. Because you know a lot of times these panels, people are paid to take surveys, so you have techniques, I'm sure, to figure out if they're just click, click, clickin'. Yes, oh yes, absolutely. And you know, there are lots of great reputable panel providers out there who work very hard to maintain very high quality panels, so you don't see that. I had a client recently, we were using a panel of C-level executives. And they said, I've been a C-level executive for 20 years and I've never taken a survey. Why would any C-level executives want to do that? And I guarantee you, it's not for these $62 that they make for taking the survey. It's because they have categories that they're interested in and they know that contributing to this type of thing helps the category. And so they want to be heard. So they do. So I don't know if there's a typical, but so obviously you've got statistics behind what you're doing. So you're a math geek, a little bit of math geek in you, right? So I mean, what kind of samples are we talking about here? Are we talking about hundreds, thousands? I mean, it doesn't have to be thousands, I presume. It doesn't have to be thousands. Minimally we're looking for four or 500 completed respondents to a survey. Substantial. Yeah, and we have studies that are much bigger than that. I am all for more data. More data is always better. But even you can get good results with small. You just can't cut the data as finely, right? You can get great. Yeah, and it just depends on what you do. You know, if you really want to parse into the difference between certain types of customers that you have or customers and prospects, then you plan for that ahead of time. You get a bigger sample of each of those so that you can dice it as small as you want to be able to dice it. So a survey of four or 500, this is pretty, I mean, high value. High value. High value, right? High value, yes, absolutely. Yeah, I've worked really hard to make it not expensive, actually. Because when I first encountered this kind of work, I was at a large consumer products company. And it was about 10 or 15 years ago. And the consumer products companies were the only ones doing this because at the time it took nine months and it cost a quarter of a million dollars. Right. So it was only the big guys who had it. So part of the reason I started my company is because I want the little guys to have it too. So I do work for very big companies. I've done work for Gap. I've done work for Tufts University, you know, all sorts of places like that. But I also do work for smaller companies that I think deserve the same insights that the big guys have. Maybe deserve more. So your cost per complete is going to vary pretty dramatically, I would imagine, but maybe not. I don't charge per complete. No? No. How do you charge? I have a flat rate. I have a flat fee for doing a project. And, you know, it depends on the client. But on average, my projects are 10 or 12 grand. And that's the same whether you have 400 respondents or 40,000 respondents. Because if you charge per complete, you're encouraging small sample size on the part of your client. And I would love to encourage large sample size. But aren't your costs a function of the number of people that you have to reach? Only if we're using a panel. Well, that's a common way to do it. But that panel, those panel fees, I price those out with the, we have got some great panel providers and I pass that through directly. I don't mark it up. Okay, so you're talking about your cost. I'm talking about my cost. Yeah, okay, well that's very, very competitive. Right, so that's the smart way to do it. You're saying, look, the list cost is the list cost. That's right. If you have a list, great. If it's a good list, we'll use it. If it's a bad list, well, that's not my problem. Well, if it's a bad list, I'll tell them not to do it. I mean, I do a lot of advising in terms of where we're going to get respondents. That's a big part of, you know, it's so funny. Everybody thinks that doing market research is all about the number crunching. The hard part of market research is setting it up properly on the front end so that you can have good data to crunch. Yeah, once you get the good data, that's when the fun begins, right? I mean, from your standpoint. Yeah, and that stuff's easy. The challenge is understanding who you need to talk to, how you need to find them, making sure that you're building a survey instrument that's getting to the answers that you want without introducing all sorts of bias, and stuff like that. And that's the hard part. Asking the right questions. Yeah, once you have the data in, quite honestly, the math we're using, I mean, it's not rocket science. There's lots of people who can do the kind of math we do. There's not a lot of people who understand the topic that we're doing and the way we do things, the way we do it. So that's differentiating for us, and honestly, that's the hard part of doing good research. Susan, I'm chuckling because we're actually in the midst of our own survey on big data that we've done at Wikibon. We've got it out there now, and I'm just laughing because of all the hard work we had to do up front. You bet. And you're right, Dave's right. The fun is when you get the data and you can start playing with it and look for the insights. Let's talk a little bit about the tools that you use. Obviously, we're here at the Tableau Customer Conference and we were talking a little bit ahead of time before our segment about some of the benefits you've gotten from Tableau. Oh, it's unbelievable. Yeah, tell us a little, walk us a little bit through the before and after story about how you kind of, what were you doing before you started working with Tableau? What prompted you to come to use Tableau and what the results of them? Okay, perfect. So our typical process, all of our projects are custom. So everything's built from the ground up for an individual customer. So it starts with a discussion of that customer, some brainstorming about some of the things they want to know. We do some research on our own. We build a survey, which obviously goes through approval processes with the client, gets fielded whether it's with their clients through an email invitation or we're using a panel. We do our statistical analysis and then the data comes out of that and that's where things have differed now that we've started using Tableau. Originally, that data would come out in an Excel file. I mean, we're a little company out. We have one seat of desktop. We're now using enterprise level data. So we get an Excel file and then I start building pivot tables, which I like Excel, I like pivot tables. I'm very grateful that pivot tables were there when I needed them, but there's no fun in doing 200 pivot tables to see what your data says. Because honestly, we're storytellers. That's what we are. Just because the data's there doesn't mean we know what it says. So it requires some work to figure out what the story is first. So that's all about the pivot tables. Then once I can sort of see sort of what the story is, then my job is to show my client what the story is. So that became a process of building sometimes 100 or more page keynote presentations of charts to walk my clients, some of whom are experienced marketers and are experienced with market research data and many of whom are not, to walk them through the results of their survey and help them understand what it's telling them and what they need to do with it without freaking them out, okay? So the process of the pivot tables and then figuring out what the story is and then building the story out in a keynote presentation could take three weeks. You know, I mean, it was just a long process. So once we started using Tableau, things changed. And the key was I needed to find a way to be able to do this work faster and bring my prices down because I had a specific group that I wanted to reach with this information. And those things were required in order to make that possible. I'm an audience person and I knew what my audience needed and I just had to find out a way to do it. But what we couldn't do was denigrate the quality of the research that we're doing. So none of that upfront stuff could change. It had to still all be done with the sort of quality controls and how we feel about doing data right and not change any of that. So now we still end up with here's the Excel sheet that's come out of our statistics package to figure out our segmentation. But now that Excel sheet gets a little bit of reshaping to fit into Tableau and it gets dumped into Tableau and everything else, me visualizing what that data is telling us, me building out a way to tell that to my client and me actually presenting the data to the client and handing it off to them, all happens in Tableau. And that piece of the process now takes me 75% less time than it used to take. So our projects used to take on average 12 to 16 weeks and now they take six. So that's a productivity improvement of, you're probably better at math than me. It's astronomical. I'm pretty significant. And even better than the time savings, because of the time savings, we've taken our prices down 60%. Wow. And product is probably better, would you say? The product's better because I'm visualizing, I'm seeing more things in the data than I can see in pivot tables. So you've actually been able to improve the quality and reduce the price. Improved the quality, dropped the timeframe by half, took the price down by 60% and we're now doing 10 times the number of projects that we used to do every month and eight times the revenue with a 60% price drop. Well, that's pretty significant. I mean, I don't know what to say about that. People are clapping here at the conference. And the great thing to me, and I have people who say, and I'm giving my session tomorrow, I'm going to talk about this. And I know there are going to be people in the room who are like, you are crazy. You did this and then you took a 60% price cut. What are you thinking? But for me, the access to the customers that I wanted to serve was critically important. And as a result of all of the incremental work that we're doing, referrals are way up, word of mouth is way up. I'm just doing tons more projects because I'm doing tons more projects. Does that make sense? So for me, it's been able to do what, in a small business, especially a service business, can be very difficult to do, which is scale. Scale is hard. Scale is a big challenge for small companies. And I've been able to take my business. I've been using Tableau for under a year and a half. And I've had this transformation. And I've added no staff. And I'm still married. And I still know the names of my children. Well, I mean, that's, you know, we talked about business is growing. And my business is on a rocket ship. And my clients are tickled pink. They love being able to see the data the way that we can show it to them in Tableau. You know, now we can, if I'm doing a presentation to a client of their data, and they say, well, I'm not sure that that's true if you look at it by UK versus US. Well, let's do that. Now I can say, well, let's look at it that way, right now. Yeah, you may be right, let's look. And I used to have to say, you know what, let me take a note and we'll do some work and we'll get back to you and we'll send you new charts in a week. And by then that issue is gone, that they, you know. Well, are we asking you again? Yeah, right. So now I can respond to those questions in the meeting with the client and put those things to risk and explore further if we need to and they love it. So our product is, you know, it sounds like Superman, but our product is better, faster, cheaper and delivering more satisfaction to our clients. And it's transformed my business. And it's because of what it's allowed us to do. And it's that thing Christian always talks about, right? Putting good visualization in the hands of, I'm a big fan of the little guy, I'm a little guy. I'm the definition of a little guy. And so while I know that it benefits big organizations from a BI standpoint, it's been transformative from my little company from a process improvement standpoint. So I'm a huge fan. Well yeah, I mean, you know, certainly we know about Tableau's got some really large customers, it's Google, Deloitte and others, but you're a two person shop and you've clearly benefited significantly from the... Two person shop this year, we'll see what happens. Yeah, well, it sounds like you're on the verge of a mega growth spurt, you got all that new revenue coming in, time to grow, right? Yeah, it's great, you know, and that's what I love about the conference. I've got the process improvement piece down. Now I'm working on how I can make this even better for my clients on the receiving end, even better for us in terms of seeing what's in the data and just sort of continuing to up our game, so. Susan, I love your story, you know, the innovation, the passing your savings onto your customers, it's great, congratulations on all your success. Thank you very much, I appreciate it. And thank you for coming into theCube and sharing your story. It's my pleasure, thank you for having me. All right, great. What time is your segment tomorrow? It's a 9.45 tomorrow morning, it's called How Tableau Helped a Small Marketing Agency Conquer the World. Well, definitely great, isn't it? It's movie night tonight, so, right? It's movie night tonight. It's movie night tonight. People aren't going to be too hungover. It's two landers on tonight, so it should put everybody in a good mood, so. All right, again, Susan, thanks for coming on. Keep it right there, everybody. Jeff Kelly and I will be back next to wrap up day two of the Tableau Customer Conference. This is theCube, we'll be right back.