Whether online or offline, the focus group is, and will continue to be a staple of the market research industry now and for years to come. Focus groups are a form of qualitative research that basically explores different subjects, perceptions, opinions, beliefs, attitudes, etc. through a guided conversation in an interactive group setting.
Focus groups, their technological methods, and what is expected of the moderators and participants (focus groupies) have really evolved over time (when's the last time you recorded a session using a VHS). Really, pretty much anything is fair game in today's focus group. I have an industry friend who's actually built outdoor sheds in her facility (notice sheds...plural) so that the shed(s) can be viewed and tested by research participants. In our Salt Lake facility, we haven't built any sheds but we have seen a WIDE potpourri of subject matter come through the doors ranging from traditional subjects like insurance, healthcare or finance products, all the way to groups around extreme sports, fish oil, beds (fully set up beds), and a myriad of other products. Really, just about any sort of product, service, or subject can be discussed and tested. And, it's a REALLY FUN research method for both focus group participants and the companies who are conducting the focus group.
I'm always watching for media about the market research industry. I ran across this video this week. Maybe you've seen it; it's a focus group with Will Ferrell, a bunch of kids, and Michelle Obama. Take a look:
Focus group moderators and organizers of focus groups really have their challenges when conducting groups. They have to keep everything organized, on track, focused, and moving forward in a way that allows companies to gather the most information possible in a one to two hour period of time. Here are some best practices on how to effectively execute your next (or first) focus group:
- Have a clearly defined purpose, and list of goals you want to accomplish in your focus group. Don't throw in the kitchen sink.
- Organize, organize, plan, plan, and plan. Don't do things at the last minute. If you're constructing a life-sized model of a car in someone's focus group facility, know how you'll get it there, how you'll set it up, and if it'll fit.
- Recruit and screen the right people, with the right characteristics, and the right knowledge. Also, make sure that you confirm that they're (the focus groupie) planning to attend (contact the participant the day before the group so you have time to replace them if you have to).
- Know your subject (or make sure your moderator is facile with the subject), but be neutral and unbiased in your conversation. Don't assume you understand what someone means because of what you know (not what they've said); if it's unclear at all, ask for clarification. No moderator comments like...that's dumb, why would you say that?
- Encourage participation by all the attendees, don't let one person dominate the conversation or someone sit in the background without a comment. Some people are more comfortable in settings like this than others; don't let that one person run over everyone else in the conversation.
- Keep things moving, don't get hung up. Focus groups encourage a natural flow of topics, don't move through the focus group topics abruptly to slowly or quickly; relax and "keep it real".
- Check the Tech. Focus group facilities most often take care of the technical aspects of the focus group for you. Check it in advance so you know everything's working the way you plan and have a conversation about how and when they'll get the deliverables to you.
This list could go on and on. Here's a focus group video (from the Academy Awards a while ago), it sort of illustrates some of these suggestions.
All in all, I strongly believe that if you plan in advance, follow the plan, and execute accordingly, you'll have great success with focus groups as a research method for your business. Focus groups, when moderated correctly, will give you a lot of information in a short amount of time in a way that will really help provide you with actionable business insights.
For all of you that have done a million of these, what are your suggestions? What would you add?
I'm really a nostalgic person. I'm not sure why. Growing up in my small Idaho town, I was SO BORED. I couldn't wait to get out of there to the big city (Salt Lake...quite the metropolis). I wouldn't want to move back "home" now, largely because of the -25 degree temperatures, but visits there definitely bring back...something...that I really enjoy. It's a little hard to explain.
I enjoy a good road trip; like to drive down Route 66 (even have my desktop background set to pictures from the highway) and see the largest vacuum museum or the biggest can of spray paint; enjoy trips to the Redwood Forest and the Gulf Stream Waters; and really seem to enjoy Disney(land) because it's flashback to an earlier and simpler era. Every era has its problems but something could be said for that time when we weren't all so...technologically connected...and we actually had to (wanted to) talk to each other.
Some of you will be surprised that I listen to country music every now and again (hated it when I was young...probably because the cowboys in my town were, well...mean...and smelly...and mean.) As an adult, I've figured out what they liked about country music. Most of you probably know who Miranda Lambert is. Her most recent song is called Automatic and stakes the claim that there was a simpler day that was better, "back before everything became Automatic." Here's the song if you're interested.
Would I go back to that simpler time? There are things that I like(d) about it...really liked about it. But I don't think I would. As Kip, from Napoleon Dynamite (one of the best movies ever filmed in rural Idaho) said, "I love technology...always and forever."
It's kind of a mix up in the mind. I'm nostalgic, but love my technology. Rather than reduce my level of nostalgia, it has made me MORE nostalgic as it's given me access to see, right away, the things I'm nostalgic for...an era...a place...a song...a person...a person at a time...all available to me almost immediately.
I was speaking to someone I met from the market research industry yesterday. We were chatting about how dramatically market research had changed in the last 15 years. He was onsite to do some focus groups and we were talking about how the process had evolved. It used to be that you'd have a list or a panel, you'd call them to do a survey to see if they qualified for the focus group (and if they wanted to participate), then you'd call them again to remind them to come to your office for the group. The group video was taped on something...VHS...mini VHS...something that seemed like it "rocked" at the time, but seems so grainy now (we're recording in HD).
Now, our groups are recruited through combinations of our panel - distributed via e-mail, online, social media, online classifieds, online ads, etc. Phone is almost the last resort. That needle in the haystack isn't as hard to find as it used to be. And, if you have to, you don't really need to get the participants all onsite because you can complete that focus group online (more to come on that subject in future posts). The audio and video is available in seconds (if you can transfer the huge file size fast enough over the internet).
This industry associate's first interaction with our company was in the early to mid-nineties. He used us to field a bunch of telephone surveys that they don't conduct any more because their customers are available online. Though there's still a place for telephone surveys (even according to him) because sometimes you can't reach people in any other way, the "easiest" method is often the best. For him, it's an online survey. The cool part of this is the continued evolution of the market research tools...past online. The online world has given us access to a whole lot of text (social media, social web, online survey open ends, etc.) that can also be evaluated (and feels qualitatively nostalgic though it's technologically driven).
If you follow this blog, you'll know I attended the combined MRA Chapter's Vegas Conference last week (may be my favorite industry conference). It was amazing how "text" focused this conference was this year. It occurred to me that these silos that we've established for ourselves in our history...quantitative vs. qualitative; full-service vs. data collection; market researcher vs. insights provider; really doesn't matter anymore...they're crumbling down around us. What does matter is our ability to answer questions in whatever way makes the most amount of sense and is the easiest. Qualitative research outputs quantitative data and vice versa, researchers are providers of answers not data collectors and full service providers, heck...we've even seen Google get into the mix with Google surveys.
I guess the point is that sometimes those nostalgic traditional research methods (like phone surveys and focus groups) are in fact the best approach to answering business questions and sometimes they aren't. Sometimes it takes a different - more technology driven tool (like text analysis and social web gathering or online communities) to answer the question, and sometimes it doesn't. As long as you have the actionable insights, the story, and the answer to your business question, the method silo or the avenue of generating the answer really doesn't matter. In fact, they're sort of like technology increasing nostalgia; they really just support each other.
Chris Robson, one of my favorite research analysts presented a very interesting presentation on text analysis. Even as a guy involved in text analysis, I learned a few things. He opened with "Time flies like an arrow, and fruit flies like a banana."...first seen on a bathroom wall. What do you focus on here? Do you focus on time flying or fruit flies?
What is text analysis? A set of linguistic, statistical, and machine learning techniques that model and structure the information content of textual sources.
We conduct text analysis on an existing corpus. A compilation or collection of text. From there we start to break things down (documents, chapters, paragraphs, sentences, etc.) Here's a typical text analysis workflow:
1) Extract content of interest / strip noise
2) Build paragraph/sentence tree
4) Remove/Code profanity
5) Identify POS (parts of speech)
6) Remove stop-words
8) Apply dictionaries
9) Code content
Text analysis is fun! It lets us view how sentences, and text is connected grammatically. Once you understand it's grammatical parts, you can start to understand it's numerical frequencies and patterns, applying math to it. Language becomes math.
Context is hugely important in text analysis. Text is messy and complex. Your ability to do this well depends on your libraries. Text analysis has become so important in some organizations that they are altering the survey research that's being done.
In the future more and more of our data will be unstructured. The tools are getting better and the techniques are becoming more widespread. It's Chris' assumption that on some level and at some point in the future, everyone will be doing text analysis.
Big data is the "word of the day." Gregory Mishkin of Market Strategies International said that it was the "thing of 2013". I do believe it will likely span past 2013 for a few years. The question he asks in this presentation is how you integrate big data with traditional research? Is it possible? The answer is yes, it's possible but it's not as easy as you think.
Here are some items and pitfalls that you need to consider when trying to blend traditional research with your big data sources:
#1) Staffing - You have to have strong marketing scientists that can analyze this data, but there is a learning curve. The data sources aren't as clean as you'd like. Find data scientists and teach them how to manage the "challenges" of big data sources. And, you'll need more people than you think.
#2) Infrastructure - Many people jumping into this space underestimate the infrastructure needs from both a hardware and software standpoint. Look out and build for the future, you'll be there faster than you think.
#3) Client Management - Don't believe the assurances. The data, when you can finally get your hands on it, will be difficult to work with. It's messier than you'll ever think it's going to be. Don't over promise because it's going to be harder than you think. What can go wrong will go wrong. Find a champion for your process...an executive champion.
#4) Project Management - Have specific milestones and don't allow projects to "creep" in scope. Without them, big data projects become far too "big". You MUST define what defines success and avoid biting off too much.
When you can bring traditional research together with the big data you can uncover things that you never would have seen or imagined before. You can build decision trees, apply heuristics that drive the trees in ways that improve the customer experience, and retain customers.
I've been waiting a long time to hear Steve Henke of Harpeth Marketing speak. He's made a name for himself teaching market research companies how to sell and market. It's no new statement that market research companies are very poor at marketing themselves. Pretty odd given the industry we're in. Here are 6 reasons that people don't have a sales and marketing process and suggestions for improving on them:
Reason #1: You're not thinking, you're just doing. You must plan and execute and follow reason #2...measure.
Reason #2: You can't manage what you don't measure. Your sales and marketing efforts must be measurable. If nothing else, use the HDYHAU (How did you hear about us).
Reason #3: You have a minimal or poorly managed sales effort. If you haven't created a sales plan; focused on hunting AND farming (business development and client management); shown the love for your largest clients; get your PM's to sell
Reason #4: You have no lead nurturing in place. Lead nurturing takes time. Marketing is about getting potential clients to know you, love you, trust you. The real hope is that you can move them through the sales funnel. Some nurturing tactics might be (email them; provide content to clients; clip articles and send them along; make phone calls; social media).
Reason #5: You don't have the money. There are a ton of free options, but you still need to plan execute and measure. Set aside some time to deal with this. Things that are easy to do, are also easy not to do. Here are some free ways to market (Content marketing, Get active on social media, become a student of the industry, include marketing on email signatures)
Reason #6: You don't really know your clients. Do you understand who your clients are when you get past the top 3? Do you understand their needs? When you look at your top-20 clients, what are the commonalities between them...if you understand those commonalities than you understand your ideal client. Are you doing client satisfaction studies or an "End of Year" survey? As an executive, pick up the phone and call a client once per week.
Reason (#7-#12): Not doing the "little things" so things fall through the cracks; Skipping "old school" marketing; Not differentiating your firm; Your messaging doesn't resonate with your audience - what makes us unique is...; Giving up too soon; Not all marketing is marketing.
If you keep doin' what you're doin'...you'll keep gettin' what you're gettin'...
For a small conference (about a 100 people) the Las Vegas Market Research Conference always seems to have some heavy hitters. The after lunch session is a market researcher from the Las Vegas Sands Corporation named Mara Riemer. The story she has to tell is about tying together the customer experience.
She opens with an interesting stat...That it only takes 2 poor touch points to lead to poor satisfaction, whereas it takes at least 3 touch points to lead to positive satisfaction. This is evidenced by the following quote, " The psychologist, Paul Rozin, an expert on disgust, observed that a single cockroach will completely wreck the appeal of a bowl of cherries, but a cherry will do nothing at all for a bowl of cockroaches. "
It's her claim that we must consider brands holistically. Consider brands from multiple touch points associated with brand...product, sensory, service, etc. They all contribute to the brand. A failure at any touch point will lead to an overall brand failure. It's my opinion that whether you are a research company, a product or service based company, whatever it is you "sell", you need to remember this point. Failure seems "easier" than success, so it's important that we focus on every piece of the brand experience.
It's interesting that often times we, as businesses, all explain ourselves in the same terms. In that case, how do we differentiate? Do we explain ourselves well so that our consumers understand our products and services? Can they find us when they know what they're looking for? The quality of your brand is always about the story consumers can tell about their experience, your brand can either feed the story positively or negatively.
It's fun for so many of us to be in this industry. The things we get to study are so interesting. We have an innate curiosity with finding things out. Riemer really appears to enjoy what she's doing, and has embraced the insights and story telling Cool Aid. It's awesome to see a group apply the craft so well.
Behavioral Economics and Neuromarketing has been under heat in the past few months. One of the presenters vocalized intents of this presentation was to help people understand what BE and Neuro consist of; that they probably shouldn't be receiving the heat that they have received; and it was mentioned that "they don't provide the answer for everything." The funny thing about that is that it seems that every new method since Online Surveys (who claimed to take over the world) has made this same assumption.
With my sociological background, I've found the medicalization of all other sciences curious. Behavioral Economics and Neuromarketing has seemed like this medicalization to the market research industry. One of the biggest insights I've picked up from this presentation is how heavily pharmaceutical companies focus on the methodology. Pharma is medical, it's marketing to medical practitioners, and I suppose it makes sense that they'd only believe the outcomes if they are neuro or medical driven. They've already been medicalized so an embracing of medical methods seems appropriate. The question I have is whether this market research method will gain traction in other areas of our industry.
Behavioral Economics methodological test results were provided in this presentation. Here are some interesting outcomes:
1. Doctors measure gain vs. loss with their patients according to how the "story" is framed, assuming that both options are safe.
2. They'd rather take small gains and improvement than a risk that no gain is made.
3. When information is framed negatively, Doctors are less likely to choose the option. When results are framed positively, they'll embrace the treatment...even though the two options are exactly the same.
It's and interesting process. I've often thought that it's important to frame negative content in as neutral or positive way as you can. This finding seems to show evidence of that approach.
People like small guaranteed gains as long as they're framed as gains.
I'm not sure if I'm convinced of the benefit of Neuromarketing, but I do seem to buy into the Behavioral Economics approach more than I have in the past. It's probably because I understand it a little more than I did sitting in on this presentation.
My favorite market research educational conference to attend is the annual MRA Combined Chapters (NW & SWMRA) Las Vegas Conference. The education is always great. The attendees are super nice, very approachable. And, the size of the conference is just big enough. The keynote for 2014 was Ben Smithee from Spych Market Analytics. Many of you know Ben. If you haven't started to follow him online than you should. He's a real thinker.
I've had the opportunity to hear from Ben before and always enjoy listening to him present. He is a grounded qualitative research, early adopter, integrator, and has excellent insights when he presents. He's a bit of a youngster (ha - I'm sure you don't feel that way Ben), and in his case it totally works. Ben really has the pulse on the Millennial generation that so many of us old folks just don't understand. He brings insights on a generation that I, too often, just shake my head at.
When he was announced as the keynote for the SWMRA conference, I completely looked forward to it and he didn't let me down in any way. Listening to Ben is like going through an appathon (marathon of App education). Every cool social app you've never heard of comes up in his presentations and it's also applied back to research...plus he uses words like "kicks" in normal sentences and it doesn't seem like an old guy trying to sound hip. If you can keep up with his break-neck speed you can learn a lot. Here are a couple of quick thoughts that I found interesting about his presentation:
1) Social has evolved since the MySpace days. Social is an attribute of the web. The social web is not about creating the conversation; it's about creating the platform or the space for conversation to occur.
2) Consumers are becoming the advocates for the brands as often (or more often) than celebrities. Your brand should drive consumers. Give them a reason to be in your store or interacting with you and your brand. Some of the best social content comes from your average girl (or guy) on the street.
3) The web has been socialized. Social is no longer a site. It's an attribute of a site or an attribute of the web. If you aren't integrating social into your brand than you're doing it wrong. There are free sources of information that help you understand social interaction and you should use them.
4) Attention, affinity acquisition, advocacy should be drivers of marketing. How do you get their attention and get them to love (or like) you? It's fine for them to like you, but how do you get them to take the next step to buy your products and advocate for you. Marketing should allow that to happen.
5) Ripple effect and speed of travel are important indicators of social content. Focus your digital marketing on what's resonating in the social web. Use it as a test.
Finally, Ben reminded us of an very important point. Understanding humans is our job as market researchers. Sometimes we forget that humans are at the center of this profession. Let's remember that it doesn't all come back to a number in an SPSS file but the color commentary that is the human experience. Well done Ben. Enjoyed the presentation!
I'm an avid watcher of the Winter Olympics. Because I'm located in Salt Lake, snow sports are interesting to me (except for Curling), and I particularly enjoy watching the Snowboarding events. I've snowboarded for many years and know how hard it is (and many times how painful and scary it is). It takes years of practice and "a ton of crazy" for the Olympic Snowboarders to do what they do. Have you ever taco'd on a rail? It hurts like $@%#. On the other hand when you hit a kicker and stomp a 360 into the pow, it feels like your riding some Knarley air. It's why Snowboarders do what they do.
Snowboarders are part of a whole different culture than the rest of us. Have you ever listened to them describe their craft? It's like listening to a different language (as evidenced above). So much so, that after a couple of Gold medal wins in snowboarding the other day by a couple of Americans, the Today Show did a piece on snowboarding lingo. Here's a quick link to the article called "Crunchy or Crispy? Al tries to learn the snowboarding lingo." A quick online search actually resulted in this more conclusive list of terms called Shreddin' the Gnar.
In chatting with some folks about text analysis recently, I was made aware (informed) that it sounded like I was speaking a different language. It occurred to me that a good proportion of the market research industry, along with the businesses that conduct market research, often don't really know what text analysis refers to and what it's capable of.
First off, text is simply the written or recorded word (for example comments, opinions shared, a speech, a social media post, a focus group transcript, etc.). Building on that, text mining and text analysis, though frequently interchanged, refer to two subtly different things. Text mining is the process of trying to uncover important information or meaning on a specific subject from text (most often large quantities of text...big data). You have a question that needs to be answered, you have text on many different subjects that relate to each other, and you try to uncover the answer to the question from the text around the subject. In essence, you are "mining" the information for the answer. Text analysis, on the other hand, is the statistical evaluation of text. It can be used in text mining but it is not a prerequisite for it. Simply put, text analysis takes unstructured text, categorizes it into understandable "buckets" with numeric values and uses those numeric buckets to analyze the text mathematically for predictive purposes or to uncover relationships.
For example, you may have a client that conducts a survey that asks participants an overall satisfaction question, followed by an unstructured question like "why did you say that". Text mining will allow you to understand the general themes of the unstructured text and gather many important insights. But, to understand how, exactly, to increase your satisfaction scores, you need to understand the predictive and relational elements associated with that unstructured text. By categorizing the responses, and then analyzing their relationships to known demographics, the satisfaction question, other survey questions, you can uncover the "cause" of the good or bad experience. From there you can incrementally alter the characteristics that emerge from the unstructured text to produce increased satisfaction with your product, service or brand.
Some time ago, we wrote a whitepaper titled "Connect with your Customer through Open Ended Context"; feel free to download it if you're interested (it's free). We believe that the analysis of unstructured text is an important piece of the market research process that is often left "undone". The Power of unstructured text is in its "spontaneous" voice that is unassociated with "structured" or "scaled" questioning. Researchers frequently ask open-ended questions but, in many cases, do very little with the information gathered because of the sheer volume of response. Whether you bucket these responses through a traditional coding process, or through a more advanced algorithmic/machine learning process, this data should not be ignored. For quantitative research, this text provides the color commentary that you really need to uncover the product, company, or brand experience. That's why text mining/text analysis is expected to show significant growth in 2014.
Have you ever had an epic fail? If you look up the definition of epic fail in Urban Dictionary, you'll find it defined as "Complete and total failure when success should have been reasonably easy to attain." My guess is that we've all had an epic fail at one point or another, or of one type or another. In my younger days, I used to do a lot more skateboarding, snowboarding, biking, rock climbing, etc. than I do now. One of my most Epic of Fails occurred on an abandoned waterslide while I was in my early twenties.
When I was young (and it was dark - so the neighbors and police wouldn't notice), we'd sneak into an old abandoned pool property that also had an abandoned waterslide. We'd hike to the top of the slide carrying an ice block and a skateboard. We'd put the tail of the skateboard on the ice block, lay down on it, and slide down the waterslide headfirst at breakneck speed. We'd use our feet for breaks, wear long sleeve pants and shirts (to avoid road rash - the waterslide had no water), and speed down the slide much faster than you'd ever move in water. To avoid the drop off at the bottom of the slide (into the pool) we'd place a flashlight at the bottom so we could see the "end was near". When we saw the flashlight, we'd put down our feet and hope for the best (more than once a friend wouldn't be able to stop quickly enough and fly out of the tube into the empty, dirty, pool).
One night, around midnight, we snuck into the slide, placed our flashlight at the bottom, hiked to the top, and I jumped into the tube (face-first). About midway down, I ran into some debris...crashing...hard (I actually ran into the piece of the slide that keeps you from flying out of the waterslide when you go around corners...the wind had blown and it had broken off into the tube). I hadn't checked the slide to see if anything was in it; because it was pitch black I didn't see it coming. I ran into the big, hard, plastic piece at about a million miles an hour headfirst. The net result, I broke four fingers (two on each hand), tore up my face, and smashed my kneecap. It hurt like crazy. In my life, I've had MANY of these experiences; broke my arm in two places swinging around in circles on a clothes line when the rope snapped; broke four more fingers trying to launch off a jet ramp skate boarding in my late 30's; got a concussion trying to do a 360 off a cliff on my snowboard without a helmet; the list goes on.
Extreme and winter sports are chuck full of epic fails. For entertainment, take a look at some really great fails. It's pretty quick.
What does all of this have to do with market research? I noticed something last week when the Denver Broncos had an epic fail in the Super Bowl (from the first snap they were never in the game). Often times, people aren't very good at predicting the outcome of a situation or event that they are emotionally or intellectually invested in. Makes sense, right?
Some of you are avid readers of this blog. Last week, we conducted some social media research on what people were saying about the outcome of the Super Bowl. When we analyzed the Twitter content from those that were predicting the outcome, 72% of them picked Denver. 72% is a HUGE amount when it comes to these sorts of things. And, Vegas also had Denver to win the game. In the past when we've run this type of analysis, these predictions have been pretty close. This one was...horribly wrong. Seattle beat Denver 43-8. Denver was BLOWN OUT. Why do you think that the public perception was so much different than the outcome?
The answer, people are emotional, and they don't have as big an impact on outcomes as they'd like to believe. The loudest person (or group of people) is not necessarily the one that is right. Additionally, in situations like this opinions don't matter much. It's the result that matters and this social conversation, though it can sway other fans, contributes very little to the actual run of play of the game. Fans can talk smack, they can make fun and berate each other, they can talk about stats, discuss the team that has the better offense or defense, but at the end of the day the game still has to be played and public opinion impacts the play very little (even though Seattle claimed to have the "12th Man."
Business is a little like that. We all claim to be the best, work very hard to be the best, but at some point someone emerges as the winner regardless of all the background noise. So what are the implications for market research and your business? Does that make market research inadequate in any way? The answer is no.
Here's a word of advice for your next market research study. When you conduct market research, you need to observe and evaluate participant and respondent behaviors and beliefs, emotions, logic, and what participants actually do or claim THEY'LL do, not what they think will be a result, or what they think OTHER PEOPLE will do. Too often companies mistake fan noise (their own opinions, their company employees, their evangelists, or their advocates) for actual consumer behaviors, opinions, actions or beliefs. When conducting research, keep this principal in mind and focus on the action the individual or consumer can control. To avoid an epic business fail, focus on measuring an accumulation of action or opinions, not a prediction of a result that is outside of consumer control. Strong consumer insights can lead to great business success...if the measurement tool matches the business question.