Marketing research can give a business a much better idea of if their current business plan is working, what tweaks might need to be made, whether they’re meeting their customers needs and what kinds of new products or services might be profitable for them. Unfortunately, many businesses don’t set aside the funds for this kind of research, or simply don’t see the long term importance of such a task.
It’s important to know how current trends are affecting your business, or what your customers actually think of you. You can take the “try, fail and learn from your mistakes” approach, or you can be more direct and simply ask your customers and prospects why they did – or didn’t – choose you.
Tom Webster is a thought leader on the topic of market research and an expert in qualitative and quantitative research techniques. He has directed hundreds of studies in nearly every major U.S. and worldwide market. HIs company, Edison Research, has the unique distinction of being the sole provider of official exit polls for every U.S caucus and election.
Rich: Tom Webster is Vice President, Strategy and Marketing of Edison Research with nearly 20 years of experience in media research, marketing strategy and product development. His duties at Edison also also include strategic development and leading Edison’s new media research efforts. He’s an expert in qualitative and quantitative research techniques, and he has directed hundreds of studies in nearly every major U.S. market and worldwide.
He’s also the cohost of The Marketing Companion podcast with Mark Schaefer – whose been on the show before – and he does a mean impersonation of Plankton from Spongebob. I was actually going to ask you, Tom, if you would do the entire interview in the voice of Plankton today.
Tom: (in Plankton voice) Of course, dear boy.
Rich: Excellent. Tom, welcome to the show. Thank you for making the time.
Tom: Thank you, handsome Rich Brooks, for having me.
Rich: I see you’re hanging out with Chris Brogan again.
Tom: I’ve learned much of my communication skills from Chris, otherwise, not so much.
Rich: So I’m just kind of curious because I’ve never been strong in math, what drew you to research and analytics? Were you always a numbers guy, counting Lincoln Logs and measuring the usage of monkey bars as you were growing up?
Tom: You have stumbled on a pet soapbox of mine. No, not at all. In fact I have a Bachelor’s degree in English, I did graduate work at Penn State in Literature and I spent a number of years as a writer and as a teacher, but the past 20 years I have been doing market research and I did go back and get a MBA. My fascination is with people but in order to understand people and do my job well, I have to have an equal facility with numbers and data as I do with language.
As a marketer, I have to tell you it irks me when another marketer says, “I’m not good with numbers.” To me that sounds exactly like, “I’m not good with words.” Innumeracy and illiteracy are things that, neither of which we can afford to be a successful marketer. So I’ve sort of made it a personal mission of mine to humanize numbers a little bit and try and take the math out of it as much as possible. But still I think to be a modern marketer you have to do both.
Rich: So in other words I’ve irked you, is what you’re saying.
Tom: I think “irk” isn’t strong enough, Rich. I think I’m enraged.
Rich: Oh man, at least vexed.
Tom: The rage is palpable.
Rich: So basically you got into the numbers to kind of understand people better so that you could help people with their marketing. Is that more or less what I’m hearing?
Tom: Yeah, absolutely. And I view it as a discipline. There is a way at Edison, my company – “the Edison way” – and we have really a zero tolerance for errors and mistakes and things that we release, and so over the years that I’ve been at Edison have been inculcated in the Edison way and that’s absolutely made me stronger at my job.
It was not my natural proclivity to be in math and numbers and statistics. Statistics is actually not a branch of math, for people that are scared of statistics. But it’s how I do my job, and how I do my job to the best of my ability is to understand people both individually and in the aggregate and stats are part of that.
Rich: Interesting. So I know that you’re very specific when it does come to numbers, in fact I think I saw a Facebook post of yours recently where you were railing against somebody for their use of pie charts. Do you remember what that post was?
Tom: I do. I’m not normally so persnickety, Rich, but this one actually did get my Irish up. Someone had posted some research that they had done on podcasting and they cited some of our research and then they said, “Well we decided to do some of our own”, so they polled their blog readers and they asked a couple of different questions in which multiple answers were accepted.
So the questions were something like not to choose the one way that you do X, it was to check all the ways that you do X. And then they presented the results for both as a pie chart. Now pie always has to add to 100, it’s a composition graph. Essentially if you had 5 or 6 items and they were all selected by 30 or 40 percent of people, that pie chart would add up to 240%, or something like that, and that’s a lot of pie.
Rich: It sounds delicious, I’m not sure what the problem is.
Tom: It’s a lot of pie.
Rich: It is a lot of pie. So I know that there is a lot of data out there, sometimes it makes sense to us and sometimes it doesn’t. I think for most small business owners and marketers, we just want simple answers to what seem like simple questions to us. Why can’t you – Tom or Edison – tell me what’s the best time to tweet or how many hashtags to get the optimal engagement on Twitter?
Tom: Because one of the biggest problems in modern marketing today is the reliance on OPD. That’s not OPP, you know me.
Rich: You’re a classy guy.
Tom: I am a classy guy. It’s OPD and that’s “other people’s data”. There’s only one way to true, sustainable, competitive advantage, and that is your own data. If you’re using somebody else’s data, then that’s data that other people have access to as well. If you read a blog post that says the best time to tweet is at 3PM and everybody starts tweeting at 3PM, that’s not a competitive advantage. That’s not your data, that’s not data that you own that’s proprietary to you that gives you a distinctive advantage over your competitors. The only path to a competitive advantage – especially if you’re trying to build audience or attract prospects – iis your own data about that audience and prospects because it’s the only thing you ever own.
Rich: Alright, so that’s really interesting, but not all of us are Edison researchers, obviously. SO if I’m a small company – whether I’m a small web design company or I own my own pizza parlor or whatever – if I can’t rely on OPD, how or what ways would you recommend that I get my own data?
Tom: Well, it depends on how valuable that data is to you. There are all kinds of do it yourself and inexpensive ways to do research – both qualitative and quantitative – and people ask me questions all the time questions like they did a survey and sent it out to their customers and they got a very bad response to it – or – not that many people filled it out – or – they got a bad response but they think that’s ok. And my response back to them is always they they provided no value to those insights, they meant nothing to you. If those insights mean something to you then there will be a value exchange between you and your customers or you and your prospects to get that information.
But we keep sending out these 20 minute surveys to people and it’s sort of like trying to get to 4th base on the first date with people on your email list. If you really value the insights then there really should be some kind of value exchange. It doesn’t have to be cash, it can be access to something behind a velvet rope, it could be a donation to charity – that works great in B2B – but every time we try to get this information and we don’t value it appropriately, we’re telling our customers that it’s really not that important to us.
And the last thing I will say about this is anytime people ask me what they should spend on market research or that they don’t want to spend money on market research my answer is invariably the same, “What’s the cost of failure?” If the cost of failure is small then don’t worry about it. Do something free or don’t do anything at all and just go ahead and make mistakes and learn from those.
If the cost of failure is really high – which is the case with a lot of our clients – then spend a percentage of that cost of failure trying to get a better answer.
Rich: That’s very interesting, and of course now I’m thinking about all of the times that I both receive these emails that say “tell us how we did”, but they don’t offer me anything, and quite honestly unless it’s something I’m passionate about I just don’t have the time to actually do that.
I love the idea of making a donation to charity for everyone who does respond to that survey, I think that’s a great idea and something that I definitely want to do.
By the way, this is not the podcast that I was planning for at all, I haven’t hit one of my questions yet, but I find this really fascinating. So if you’re ok with this, I’m just going to ask more questions about market research.
Tom: That’s fine. I know a little about it, but I’m fine going any way you want to go.
Rich: Alright. Well you mentioned making a budget for market research, but even before we get there, let’s talk about surveys for a second. So we get them all the time, whether in person or we go to a brick and mortar shop or because we have subscribed to some social media guru’s email list and they want to know more about us. Is there a type of survey that is bound to get better results because of the questions we ask or the way it’s presented? Have you guys done any research into that or do you have any insights into that?
Tom: Yeah, we do a lot of research and have a lot of insights into that. It’s all about the response rate, right. It’s less about sample size than it is about sample quality. If I show you a research study that has a sample size of 43 people and I say this represents marketers, I’m not doing very good research here, unless those 43 people are 43 of the Fortune 100 CMOs. Then you’d have a very different perspective on that. So it isn’t necessarily about numbers as it is how representative it is of the population that you’re interested ins and response rate and obviously the quality of the questions and the quality of the data.
But just to get back to that concept of a value exchange, I think one of the most important questions that you can ever ask at the time of someone purchasing your product or service or even requesting more information about your product or service, a lot of people ask, “where did you hear about us?”, and there is some value in knowing that. A better question is just, “why us, why did you come to us, why did you buy from us?” Make that a really simple one question with a fill in the blank answer and give somebody some incentive to provide you with that data, and it has to be bran appropriate. If you’re Tiffany, that’s not a 10% discount, but it might be access to something else. Every brand has their own way to make customers feel special, and as long as you don’t violate your brand promise, you do that.
Instead of going out and saying we need you to complete this 20 minute survey, try to make little value exchanges multiples times, It’s a relationship and the information you get from a long time customer is more valuable than information you might get from a one time customer. So string that out, have a relationship, and the more a customer trusts you and does business with you, the more information they’re going to share. And if you’re trying to get information from people who don’t trust you and don’t want to give you that information, then that’s the kind of information you do have to pay for. You’ll have to sort of go out and get that through other means.
Rich: So sometimes we can go to our audience, to our community, to our customers and ask them a question. Other times we may just have to pay for a market research firm to do that kind for work for us.
Tom: Well, try to spend some money on talking to a different sample than you’re used to talking to, and whether that’s through a market research firm or doing it yourself. If I send out a survey to people who read my blog or listen to our podcast and ask them to give their opinion of the podcast or blog and they already kind of like it – or else they would read it or listen to it – then there is already that bias already.
Then there’s the bias of people compelled to respond to that sort of thing. So there’s nothing really representative about doing that kind of research. It can give you some good questions to ask and answer in other ways, but a good rule of thumb here is in order to really know something you kind of need to know your data sets. You need one data set to give you a really good question.
And let me clarify that by going back to the “what’s the best time to tweet” thing. Let’s say that you mine your data and you see the tweets that you send out at 3PM seem to do really well and you have corrected for the fact that maybe you send more tweets at 3PM – and a host of other variables. What that does not give you is an answer. What that gives you is a really good question, and that question is, “I wonder if time of day has any bearing on the effectiveness of my tweet.” You’re given a piece of random data, there’s a spike – it could be random, it could be meaningful – the answer for you is to then find another data set, query that data set somehow and find out if there is even a relationship between time of day and the success of the tweet. But if you dredge your data for points like this, you just might easily be finding random, spurious correlations.
Rich: So if I understand you properly, I may be looking at my tweets and seeing I get more engagement around 3PM, but that may not be the time of day. It may be that I used a hashtag or that I was especially funny at 3PM in the afternoon and I get better engagement.
Tom: Yeah. The thing is you don’t know why to any of that, you simply have an observation. And that observation is that in the past these tweets are sent at 3PM – with other information – have done really well, and what that should do does not compel you to tweet at 3PM all the tie. What that should do is compel you to ask some really good questions about that like, “I wonder if time of day has anything to do with it.”
You look at cancer clusters in this country If you plot recorded cancer cases on a map, you will see clusters. And that leads you to believe it may be cancerous to live in certain cities. Well that may not be true, everything clusters, random numbers clump and are not evenly distributed or they wouldn’t be random. So there’s going to be some time of day in which your tweets are more successful than others, but that might not have anything to do with time.
Rich: Interesting. You said there should be 2 data sets we should be looking into, what are the names of those data sets?
Tom: It’s not that there’s names, you should just try to find another source of data. So for instance, if in your Google Analytics you notice that certain types of content are doing better, then that’s a good time to go ask your audience a few questions to test a hypothesis. Let’s say your hypothesis is that content that includes a cat does very well. Just dredging your data and seeing that gives you a really good question, but just seeing that could be random and spurious. So instead ask some people and get their opinion on it.
The problem is when we dredge data and take things out that appear to us compelling, it can lead us into what I call the “optimization trap”, and I think this is crucial to today’s marketers. So many of today’s marketers are hyper focused on optimization and they don’t really understand the big picture of what they’re doing.
Let’s say your emails get more opened at 3PM, so you start sending all your emails at 3PM and you think you’re doing the right thing, you’re optimizing. But let’s say within that spike there are still a ton of emails that get opened those other times of the day. So now what you have done is you have hyper optimized on this one chunk of people who seem to be more responsive at the expense of all these other people. And then let’s say that you ask those people questions and they respond and you hyper optimize on that.
What happens is you keep focusing on an ever shrinking target that may or may not be at the center of what your brand is really for. So if you don’t have a sense of what the broad circle of the audience of your brand looks like before you start hyper focusing in on this and that, you could be cutting back wide swathes of potential prospects and customers for the lure of easy optimization.
Rich: That’s fascinating. But I guess my question would be again if I realize that my emails are being opened at 3PM, and the correct answer isn’t to send out all my emails at 3PM, is there a better solution out there?
Tom: Well, you need to figure out why the emails that are opened at 3PM are opened at 3PM. It might be that it’s idiosyncratic to that particular culture, or it just might be random. And if it is just random, then don’t worry about it.
Rich: Alright, I’m going to stop worrying about it. So one question that I get a lot from my clients – a lot of small businesses – when I sit down and talk to them is, “where is my audience online?” Is there anything that our research can tell us whether we have to go out and reach a new audience or serving our current audience, what are some tactics that you might have where we can really identify where our ideal customer – our business avatar – is hanging out?
Tom: Well, I will tell you the central problem with the question you just asked me is the use of 2 words in the same sentence, and that is “audience” and “customer.”
Tom: The customer is one thing. You know how to find your customer, there is a finite set of tools to find your customers who are in the decision process. But then there’s this whole other universe of people that have all kinds of varying possible relationships to your brand that would not be in this database.
I’ll give you an example of that. A number of years ago, GlaxoSmithKline was going to launch a drug for restless leg syndrome. On the one hand you could say their customers are people that have been diagnosed with restless leg syndrome. However, nobody knew what the heck restless leg syndrome was. For a lot of people it was just an inconvenience, it’s not a disease. It might just mean you get up a couple times at night and never think about going to the doctor for it. So they had to do all of this advertising just to teach people what restless leg syndrome was just to get people to go to the doctor. That’s understanding an audience. Understanding a customer is saying if you have RLS you need this molecule. Understanding an audience is realizing there are people out there who don’t even realize they have the problem your product or service solves. There are people out there who do know your product or service but have rejected it for reasons you may or may not know.
The ultimate questions – and I submit this is the most important question that you answer as a marketer – is this, “why or why not”. Why did they buy or not buy your product? When we focus on customers and people who are engaged in our content, we’re focused on a portion of our audience. What we’re missing is the audience that doesn’t know who we are, or that knows we exist but despises us, and they’re all part of that stew.
Rich: Is it that I formed the question poorly or is it that there’s just no simple answer to where is my audience?
Tom: There’s no simple answer to where your audience is and there never will be. There’s no perfect study I can design that will bring you to your audience. I can get you closer, but it is a difficult thing to do. Every brand has the thing they think they stand for and then they have the thing that other people think they stand for. And you can have a decent idea of what your regular customers think you stand for, but that may not be what other people think you stand for.
Take a brand like Taco Bell, I think they understand their brand very, very well. When Taco Bell launched in this country, they launched with the brand that they were the fresh alternative with lots of knives cutting through glistening lettuce on the early commercials. So that’s what they thought they were. And eventually they realized what other people thought they were, and Taco Bell I think has become this sort of wonderful brand for people who are stoned and up late. But that’s because they went out and understood who are these people who are buying all our tacos. And that’s the smartest thing you can do as a brand is not force feed all your marketing into that bucket. Figure out why you’re having that success and then create an exportable insight to other parts of your business and use that knowledge to grow in other areas.
Rich: So it seems like the most important question we can be asking ourselves and perhaps our audience is why. Why us, why are you choosing us, and that may lead us is not to the promised land, at least point us in the right direction.
Tom: Absolutely. Why us or why not us. Why wouldn’t you buy from us, why don’t you consider us, or why do you? It is easier to get the “why do you” information than the “why don’t you”. The “why don’t you” information people will tell me that why that’s too expensive or why I don’t need it. And to them I say you’re right, only the very big companies do that.
Rich: You’ve got an answer for everything. Alright, one last question and then I’m going to ask you where we can learn more about you online. Do you have any tips or tricks or strategies for getting people at a conference to actually fill out the speaker evaluations? Should it be on paper, should it be electronic, should I not let them out of the room until they tell me what they think about the speaker? I’m really at a loss.
Tom: Well, I will tell you what my company is best known for – even though I do speak a lot of new media and social and behavior and things like that – my company’s single biggest job is we’re the sole providers of the exit polls during national elections, primaries and caucuses. So if you ever watched television on election night in 2012, all of the exit poll data that you saw no matter what channel you watched came from us, we’re the sole providers of that. We have been for 12-13 years now. I can tell you that if you station people right outside the room with a 1 or 2 question response needed on a clipboard you’ll get 40-50% response. Just be nice about it and make sure that there’s plenty of time for people to do that.
Rich: I’m going to see if I can work that in this year. Tom, this has been great and I really appreciate you rolling with the questions because really they were just coming up out of nowhere in my head and you did a great job talking about them. Where can we find you online, where would you like us?
Tom: Our work is at Edisonresearch.com. There’s plenty of stuff that we’re putting out there in terms of research I have a Twitter account @webby2001, and I have a blog at BrandSavant, but I don’t write very often.
Rich: Alright, sounds good. We’ll have links to all that in the shownotes. Tom, once again, thank you very much for your time today.
Tom: My pleasure, thank you, Rich.
- Check out Edison Research’s website to find out more about how Tom and his company help businesses grow by conducting market research techniques.
- Follow Tom on Twitter and be sure to check out his blog and informative podcast!
- It’s not too late to get your tickets to the Agents Of Change Digital Marketing Conference! Get a front row seat to some of the biggest names and thought leaders on search, social and mobile marketing. Can’t make it in person? No worries, digital tickets also for sale!
- Transcription services provided by Jennifer Scholz Transcription Services.