Like many other Canadians, I received a card in the mail from the Government of Canada promoting a website named MyDemocracy.ca. Just a day before, I’d also come across a link for it on Twitter so with two hints at hand, I decided to read the documentation and find out what it was all about. Along the way, I noticed a lot of controversy about the survey so I thought I’d share a few of my own comments here. I have no vested interested in either party. I am simply a fan of surveys and have some experience in that regard.
First, let’s recognize that one of the main reasons researchers conduct surveys is to generate results which can be generalized to a specific population, for example the population of Canada. Having heard of numerous important elections around the world recently, we’ve become attuned to polling research which attempts to predict election and electoral winners. The polling industry has taken a lot of heat regarding perceived levels of low accuracy lately and people are paying close attention.
Sometimes, however, the purpose of a survey is not to generalize to a population, but rather to gather information so as to be more informed about a population. Thus, you may not intend to learn whether 10% of people believe A and 30% believe B, but rather that there is a significant proportion of people who believe A or B or C or D. These types of surveys don’t necessarily focus on probability or random sampling, but rather on gathering a broad spectrum of opinions and understanding how they relate to each other. In other cases, the purpose of a survey to generate discussion and engagement, to allow people to better understand themselves and other people, and to think about important issues using a fair and balanced baseline that everyone can relate to.
The FAQ associated with MyDemocracy.ca explains the purpose of the survey in just this manner – to foster engagement. It explains that the experimental portion of the survey used a census balanced sample of Canadians, and that the current intention of the survey is to help Canadians understand where they sit in relation to their fellow citizens. I didn’t see any intention for the online results to be used in a predictive way.
I saw some complaints that the questions are biased or unfair. Having completed the survey two and a half times myself, I do see that the questions are pointed and controversial. Some of the choices are extremely difficult to make. To me, however, the questions seem no different than what a constituent might be actually be asked to consider and there are no easy answers in politics. Every decision comes with side-effects, some bad, some horrid. So while I didn’t like the content of some of the questions and I didn’t like the bad outcomes associated with them, I could understand the complexity and the reasoning behind them. In fact, I even noticed a number of question design practices that could be used in analysis for data quality purposes. In my personal opinion, the questions are reasonable.
I’m positive you noticed that I answered the survey more than twice. Most surveys do not allow this but if the survey was launched purely for engagement and discussion rather than prediction purposes, then response duplication is not an issue. From what I see, the survey (assuming it was developed with psychometric precision as the FAQ and methodology describe) is a tool similar to any psychological tool whether personality test, intelligence test, reading test, or otherwise. You can respond to the questions as often as you wish and see whether your opinions or skills change over time. Given what is stated in the FAQ, duplication has little bearing on the intent of the survey.
One researcher’s opinion.
Since you’re here, let me plug my new book on questionnaire design! It makes a great gift for toddlers and grandmas who want to work with better survey data!
People Aren’t Robots: A practical guide to the psychology and technique of questionnaire design
Live notetaking at #IMD16 in chicago. Any errors are my own.
Marketing for MR: What we’ve learned from GRIT, Our clients, and our own marketing by Lauren Tilden and Lukas Pospichal
- GreenBook mission is to connect researchers regardless of the size of the business
- What does GRIT tell us – GreenBook research industry trends, it’s for researchers, tell us where we are and where we are going, not being used to inform marketing for research company
- How do clients choose suppliers? [he shows a hurried tiny bar chart and laughs 🙂 ]
- Stated importance on GRIT says that relationships matter first, price is last
- Radio landscape map analysis chart – clustered different criteria for selecting suppliers, quality, experience, consultative skills
- For small budgets, price is important. But as you move to higher budget its more about quality and consultativeness
- Top 5 sources – seminars or conferences, industry websites, face to face events, webinars
- Clients don’t want to be sold to the second you meet them, get to know them first
- Find out what the company does first before you meet them
- Clients say – I want my research to change the decisions of my marketing executives
- Help your clients promote, distribute and present your research internally
- Develop templates, resources, processes for them, help them jointly deliver the results to other teams
- Good marketers have a plan, good marketers go easy on the sell, the second you start a. sales pitch on your webinar people drop out, good marketers have leaders who provide resources and staff, good marketers go where their clients are, not necessarily market research conferences, good marketers experiment
- Provide value and use it as a lead generation tool,
- Messaging should focus on a specific person, use real words, eliminate needless words, think about what you want people to do
- Tuesdays and Fridays at 9am work the best for them, figure out what times work best for your clients
- Use words like you, you’re, your, create a sense of urgency with words like last chance, or hint to a gift or something free
- Effective webinars are hot buttons or broadly interesting to many people
- Teach people something don’t sell, unless the webinar is advertised as a product demo or similar
- Be unique, or uniquely good
- Market the market research industry , don’t focus on the features of your company but rather on the benefits
- Yes or No how
- They get vendor recommendations from colleagues internally, check greenbook or quirks, or see who did a good piece of research, conference talks are a good place to find new ideas, watching webinars because there is not travel, vendors can recommend someone they trust maybe the person they lean on when they cant take a job
- Partner is someone they go back to again and again, give a five year contract without thinking twice, they will keep a spot open while waiting for a partner; a one time project is a vendor
- How does a vendor get to partner status, requires trust, lots of responsibility, by the tie vetting is completely done its almost like they are partners by then
- Like webinars with case studies, with companies they know and recognize, don’t really look for blogs and not fishing for content
- Emails without a hard sell are more compelling, want to hear about competitors
- Know what youre really good at, be aware of the clients business problems or ask them outright and describe your options for responding to it
- Bring a methodologist with you to a capabilities present, take the phone away because the client is more important than your phone
Live note taking at #IMD16 in chicago. Any errors are my own.
Panel: Strategies of successful research agencies, Gillian Carter, Ross McLeanr, and Arusha Sthanunsathan moderated by Lukas Pospichal
- Clients don’t know what they’re buying until they are fully on board
- Use client’s excitement to book speaking engagements, win win for them to shine among peers and the research company can share their expertise through the lense of a client
- Overcommunication helps to avoid problems, overshare until you’re told not to
- Daniel Kahneman – experience is measured by most intense positive, negatives, and the end, and these are averaged for an event, measure these points well
- Use the advantages available to you, whether you are small or large, stand up for what you believe in, smaller companies can react more quickly
- Best clients will often let you talk about them in sales meetings even if they don’t want you doing so at conferences
- Leverage client pride in your projects, find all the spaces where their work deserves to be showcased and help them become more publicly recognized, and hey mind doing a case study for us?
- Work hard to make your clients look smart to their superiors
- She gets big respect for being able to say the alphabet backwards really fast
- Should my company be a pokestop? Is this for business, what is my strategy? What is the right question to ask?
- You should be asking how do i do this.
- 200 million numbers are on the do not call list and 44% of direct mail is never opened [me and me]
- People aren’t watching commercials either so should we move commercials over to where people are watching now?
- The vehicle/channel is being discarded, consumer behavior is changing
- People watching changing behavior will win, if you uncover meaning in your own behavior you will win, create epic content and you will win
- Companies have changed from big media buys to social media buys but they haven’t changed what they’re offering
- People won’t tolerate impersonal messages anymore
- Consumers will no long tolerate companies that inconvenience you, “batteries not included” is no longer acceptable
- Make sure you can get to your own data, you need meaning of this data
- We don’t help our clients understand the outside world enough, we focus too much on inside data
- Your goal isn’t more facebook engagement, your goal is more clients. Potential clients need to find your facebook page, click on your fb CTA, and proceed down the sales funnel
- ABC – Always be closing, ABH – Always be helpful, is your service helping to make their day better
- You need to put your top people on content marketing, it’s not a job for interns [oh my, the worst blog posts come from people who are trying to fill word counts not create opinions]
- We let social media take us wherever it wants to go but you must have a strategy
Branding you: Sales tips for market researchers by Dan Rangel, Survox
- Join a few meetup groups, and maybe start your own, then you’re in a leadership role
- Consider putting your photo on your business cards
- Althways think about WHY should this person do business with me
- Show them the money, talk about ROI
- It’s not about you, always listen.
- Weekly project plans are important for the larger projects, let client see where the status is, what they will need to do, what you still need to do
- Nurture the human bond. Go to a baseball game, lots of fun, lots of talking time, and builds a good relationship
Live notetakeing at the #IMD16 conference in chicago. Any errors are my own.
No more eblasts: reimagining email for the modern subscriber by Monica Montesa, Aweber
- Email is not dead, it’s evolved into bigger and better
- Email shouldn’t be measured by how many subscribers you have
- Email is not just about making a sale
- IT’s time to embrace the human on the other end of the email
- Personalized email get a higher click though, 40% higher
- Not allowed to use the word eblast, feels one way and self serving, makes content seem like it’s for no one in particular
- Email is for more than just company news
- People prefer email communications over social media, but you must deliver value
- Broadcast email is a one time notice, maybe time sensitive, maybe promotions or discounts, share blog content, maybe include some user generated content like client stories
- Emails don’t have to be a sales pitch every time, newsletter can remind that you are a thought leader
- Consider auto-responders, welcome series that triggers for people who just signed up, include evergreen content with no time deadlines, include introduction to you and company, an ebook, contact information, an educational course
- A course doesn’t need to be a huge thing, maybe five emails positioned together
- Aim for quality not quantity for an email list
- List building tactics – set proper expectations, make sure signup form describes the content they will receive, engaging call to action, avoid boring words, offer an incentive to signup like an ebook or checklist or a free consultation
- Meet your audience where they are, promote your list on your social channels so you can control who sees it, consider ads to drive traffic to the signups
- Ebooks give a lot of value, Balance value with promotion
- Consideration – introduce your product as a solution
- Conversion – convince audience to sign up, don’t be shy about the sale, share testimonials, discounts and consultations work here
Putting your email marketing to work: generating and prequalifying leads at scale, by Ana Jacobsen, Drip from Leadpages
- Are you happy with your op-in rate?
- List growth is critical to business growth
- Is the opt-in on your website hidden? Do people have to hunt for it?
- Do you ONLY collect emails for your newsletter?
- It needs be on the homepage along with something valuable
- Welcome mat is very important, tell people why to opt-in, spotify uses it to generate users growth, uber takes over the first page for it
- SumoMe is free and recommended
- Pop-up or widget is also effective, can be irritating but done well can engage folks. Pinterest does this. L’Oreal does it also but they don’t say sign up for our list, they say sign up for free samples
- Landing page must stand alone and convince them to convert, no nav bars, no footer, no chat box, just design to get conversion – CARFAX, Oprah website – if there is only one button on the page it will get clicked
- Least obtrusive is the top bar on the website – HelloBar is free, works well on WordPress
- But an invite on blog posts, you know they have prequalified themselves as most interested, maybe even match the blog message to the invite message – DrIpForm
- How many white papers are on your hard drive, Ryan Dice is a great marketer, recommend following him, but how any of these white papers have you actually read, are people actually engaging with them?
- The case for campaigns – follow up every white paper download and see if people liked it, wanted to commend on it, or questions, follow up for a 4 week time period
- Subscribers should not get generic followups, interested visitors will fill out lots of data, can push people towards the right email followup
- Emails can link to your calendar where they can choose a time to speak with you, reminders stop for them but not for other people
Contented: Learn to love the art of creating relevant stuff by Susan Griffin, Brainjuicer
- Need to consider what you want to say and what clients want to hear
- Creating and sharing an asset that is helpful and informative, content has to tell your story, what makes your company te best resource
- Snail mail now email, travel books are now travel websites
- Need to apply your principles to your clients and yourselves
- Fluency – distinct assets used to recognize a brand act as a toolkit to build market share
- Fame – how readily a brand coe to mind predicts market share
- Have many touch pains – website, case studies, white papers, get individual items that work together
- Mark Earls, Herdmeister, a thought leader around social, we dont do anything by ourselves, light little fires and one of them start a big fire
- You can say you do conjoint and segmentation but you’re basically telling clients you sell wrenches and screwdrivers, you need to say you understand their needs, those tools aren’t distinctive
- Don’t be narcissistic brand, use product names that people recognize
- Marketoonist is great at this, cartoons are $35 and they are powerful story tellers
- If you’re going to do social media, do it seriously, don’t be the person with 8 followers and 3 tweets
- Figure out how to blog regularly even if it’s just a tiny share
- Relevant content comes in different sizes – a tweet, a white paper
- Recycling is good, you can share old articles again, repurpose the content
- Don’t scare up personalization, get the name right or don’t use the name at all, make sure first and last names aren’t flipped
- Content needs to grab people, don’t be bait and switch, be bait and catch
- A How-To gives readers a quick thing to read of just 5 actionable points that is relevant to a problem
Live notetaking at the IMD16 conference in Chicago. Any errors are my own.
- Don’t go home and begin the content, go home and start planning first
- Some people in this room don’t have a marketing plan for this year
- Developed a marketing and sales period which is a 7 step process
- Start by doing your homework, understand business environment, decide on your strategies which set the tone for the rest, build awareness, generate leader, nurture those lead, create first time clients, create repeat clients
- Homework tips – 3Cs, company, client, competitors – financial analysis and understand where your revenue is coming from, do a SWOT analysis, do post-project surveys for client satisfaction, projects are the reason they keep coming back to you
- Pareto principle 20/80 rule works, who are your 20% of clients and what is the commonality, size geographic industry, where are the consistencies, do a year end survey, why did they hire you, what is unique compared to competitors, do they use other companies and why
- Visit competitors website a few times per year, their services and employees, try some secret shopping
- 2Is – industry.= what trends will impact the business, new technology, new players
- A – audit, find somebody outside firm to do a marketing and sales audience, have them review plans, reports, proposals, website, get feedback from someone who doesn’t stare at it everyday
- Strategy tips – most important part of plan, right strategy will have some success even if tactics aren’t great
- Know who your ideal buyer is, ideal company you want to sell to, define them clearly to create your plan
- Points of differentialion, how do you want to be perceived, everybody can’t be great and have great employees, that’s the cost of business
- Are there opportunities, should you pay attention to big data, at least be aware and consider it, are there problems in the industry or my company that need fixing and will get you far ahead, sometimes education can fix a lot
- Two kinds of clients – new clients and repeat clients, allocate funds to marketing to both types
- Work on SMART goals, you can’t just set a goal and expect it to happen, you need to outline activities that help you achieve those goals
- Growth grid – current vs new products and current clients vs new clients. Core business is current clients and current products.
- Awareness – website is the core of your marketing, well written, attractive; go social via linkedin twitter facebook google plus
- Email marketing – get people to know who you are, stay top of mind, email does this, PPC/SEO/advertising
- Network and exhibiting need to be added too, people to people environment is also necessary
- Don’t have a LinkedIn page, have a linkedin presence
- He’s not a believer in cold calls but do what works for you, if you do a good job, you only have to follow up on leads
- Good content builds awareness and generates sales leads, give some away through blogs, but ebooks and white paper downloads are emailable leads
- Webchat on your website costs 20$ a month, can have somebody ready to to chart at any time, lets you see when someone is visitin gand you can jump in right away, don’t expect huge return on it
- Network and exhibiting are the best, his notebook is the best, he writes down everyname of who he spoke to and what they spoke about, he always writes notes down, even what someone’s hobbies are
- Linkedin is a quality play, he turns down many invitation, only accepts people that there is a chance with, you can export to email list [careful, in canada this could get your in ltrouble]
- We often do a bad job of working a booth
- Get contacts to introduce you to people, their boss and colleagues, and stay in touch with those people, if your one contact leaves then you’re not in trouble
- Nurture people until they are ready to buy
- Email is to connect people with content, not to sell. Give them something of value, a blog post.
- Participate in linkedin groups, that’s where your clients and prospects are
- Lead nurturing – don’t take everything digital or remote, pick up the phone, go to conferences early and meet with clients
- Share content by email – i say this email and though you’d like it. Cheers. A really short email. Set up google alerts to find this content.
- First purchase is the hardest purchase. Please ignore the people you’ve worked with forever and take a chance on me. Need to mitigate this fear.
- Have a point of differentiation. Have a proof sources and case studies, clients and logos, white papers. LEt them see they aren’t your first client. Let them see your are knowledgeble enough to write about it. DO you have a first client deal?
- First time clients – capabilities presentations and proposals are way to long and me focused. They already know what you do and you don’t need to tell them. YOu don’t need to say how awesome you are. You need to tell them what you can do for them. What’s in it for them. Throw a one page summary of your company at the end of the proposal.
- Stay in touch with clients by email, share content, be active, participate in their conversations. Stay top of mind.
- Project followup – be a client advocate who talks to clients separate from the direct client relationship. Be the problem solver
- Send a handwritten thank you note. Don’t save the gift basket for the holiday season.
The Brains Behind Better Marketing: Using Neuroscience to diagnose and optimize marketing efforts, By Michael Smith, Nielsen Consumer Neuroscience
Live note taking at the November 10, 2016 webinar. Any errors are my own.
- Consumer neuroscience is relative new, over the last decade
- Think then feel, weighs pros and cons of product benefits and through rational optimization and then you think about some products being more valuable. This is completely backwards. Maybe we use emotions as a first gauge prior to coming to thinking decisions.
- This is not new, Kahneman wrote about system 1 and system 2 which are thinking fast and thinking slow. Fast is intuition, automatic, and emotional. Slow is emotional, deliberate, and logical. Hare and turtle.
- System 1 starts before system 2 is even on board.
- Traditional consumer insights are market data, POS, panel, explicit data, focus groups and questionnaires. We also need implicit, non-conscious, and physiological reactions to get a more complete view of the consumer.
- Tools include EEG, core biometrics, facial coding, eye tracking, self report
- EEG – 32 sensors collect data 500 times per second to capture activity across the brain, can measure response to marketing materials
- Biometrics – galvanic skin response/sweat, sensors on fingertips, heartrate
- Facial expressions to show surprise, confusion, joy, sadness, cameras also show where eyes are looking at an ad or commercial
- EEG trace has a lot of granularity, change it into a profile of activity over time, aggregate data over many people, can see high and low points, which scenes are high or low engaging [wish he’d talk about people not consumers]
- Can measure memory activation, attention processing, and emotional motivation
- Have 80 years of research on this so we know what is getting into memory, degree to which people are engaged in the communication, and intentional attention and processing
- Biometrics give us momentary engagement, degree of arousal from the ad, emotional highpoint, does it grow over time or finish on a strong point, do they tune out before your branding occurs
- Facial coding and expressed emotions, if there is no emotion the ad won’t be successful, impacts success of delivery, some ads are designed to create a specific emotion
- People are drawn to the center of a stimulus and they naturally attend to faces, people want to look at people, but you might want people to look at your 1-800 number or your logo
- Neuroscience tools are predictive of sales – neuro combined is the best, followed by EEG, biometrics, surveys, and lastly facial coding [of course, the best tool is always a combination of tools]
- Case study – public service advertising – Cheerleader PSA, ad to encourage dad’s to be involved in their kid’s lives – Woman is upset about a crazy man dancing outside her window but then you see he is cheering with his daughter
- People liked the ad, 79% top 2 box
- EEG and biometrics scored it high, lots of engagement, attention, engagement
- There are peaks and valleys at various parts of the 30 second ad, at the lady scowling, seeing the cute little girl
- Biometric trace shows a slow build, had a positive call to action
- Put EEG, biometrics and facial coding together on one chart, kind of neat, negative expressed emotion at the beginning but becomes very positive at the end
- Heatmap shows ‘attention vampires’ – people are looking at irrelevant things over the logo and phone number, it’s nice to look at the little girl but you need people to see your brand, maybe put the call in number to where people are looking and reduce the distractions when you show the important info
- Neuro measures the non-conscious, ensures emotional connections exist, provides granular diagnostics
- If you have norms, do you still need neuro? quant alone is only part of the answer
Live note-taking at the November 10, 2016, webinar. Any errors are my own.
- People aren’t just buying products or services, they are trying to get things done, many ways to get these things done
- How can we sell more ice cream? No calories, better distribution. Now think about you personally the last time you had ice cream, why did you have it? Completely different answers, nice way to end the evening, bribe the kids, want to cool down after a run, playgrounds, coffee, cigarettes, beer are the competition. It’s not an ice-cream company and the competition is much broader.
- Can do job quicker and faster because you’re more targeted on what needs to be done
- Need to know why consumers prioritize certain jobs and why they buy the things they buy
- When: understand target markets fully and meet the needs of multiple stakeholders
- How: address root causes and specific use cases
- Why: broad solution space base don both functional and emotional insights
- Discover the jobs, what do stakeholders want to accomplish, what are the pain points and current approaches but don’t start there are you will miss the why
- Identify the success criteria and investigate the obstacles, go into detailed planning
- Assess the value and beat the competition, do NOT start by trying to beat the competition
- Credit card company – are you in the ‘gold card’ segment? Not very sophisticated. Add demographics which isn’t much better. Maybe go further with ten life-stage segments (college students, couple with baby) but this is not all that predictive for a gold credit card. Company really wanted to attract high spenders – business travelers, small business owners who put business charges on personal cards, these are very different people. Not everyone is seeking more even though that’s what we’re trying to give them. Business travelers might want instead preferential events so they can connect with their family at home.
- Toothpaste – satisfies a very specific job. Is Wisp the worst toothbrush? It’s hardly a toothbrush at all. But it’s a huge category success. Has a very clear view of the customer. People who travel during the day and it’s hard to brush your teeth when you’re away, don’t want to carry a wet toothbrush. You don’t need a dental hygiene cleaning, you need to make sure there’s no broccoli in your teeth. Can’t be bulky. Have to be okay with throwing it away. [Darn, they almost had me until I heard it’s disposable]
- Décor Aid redefined the norm by focusing on hundreds or low thousands of dollars clients, not hundreds of thousands. Value of freshening a living room before a party. Give you a newly interesting den. No obvious competition other than DIY and maybe it looking like DIY.
- Go from big to narrow, get more targeted over time.
- Case study – prepared food company expanding product portfolio. Functional job – budget, nourish, reduce time on menial tasks, eat healthy. Emotional job – unwind, bond, try new things, variety, demonstrate caring, express creativity. Makes more sense to start with the emotional jobs in this case – “feed yourself” isn’t very useful. Success criteria – tasty, filling, healthy, easy, fresh, variety. Obstacles – lack of energy, lack of ideas, distractions, cravings, timing. There are industry benchmarks and you have to be prepared to break those so you can stand alone a bit. Myth – couples want quick meals is more correctly couples want easy meals.
- Jobs To Be Done goes different than needs or features. It’s only one piece of understanding the market. Don’t compete on features.
- Map the entire process of customer behaviour and keep digging to get a root causes. Be a kindergartner and keep asking why why why. Create a hierarchy of stakeholder jobs and priorities. Satisfy both functional and emotional jobs.
- Get out of the conference room and talk to actual people. people have a hard time telling you these kinds of things on a questionnaire. You can quantify context and experience, you can ask about frequencies. And then qualitative gives you the full experience
- Two segments is probably too simplistic and too many is unreasonable. Direct mail gives you more options for more segments. 4 to 7 is probably a good balance to exploit richness of the world but not getting over detailed.
What Customers Will Want: The How-To of identifying the future of customers needs and wants by Greg Yezersky, Oakland University
Live note-taking of the November 10, 2016 webinar. Any errors are my own.
- Average company lifespan on S&P was 65 years 50 years ago. Now, projections say it is about 18 years.
- 86% of original fortune 500 companies are gone; of fortune 100, 43% are gone
- Permanently excellent companies and industries do not exist
- Challenge: what is the root cause of success
- Research, concept, design, production, sales, all of this takes time. What people wanted before, they might not want anymore. We need to know what people will want in the future. We don’t have a crystal ball.
- Traditional approach is to follow trends, square shoulders and holes in jeans leads to new fashion trends so we adjust are machines for those, and then the trend goes away
- People don’t know what they will want in the future
- Don’t identify the right problem, identify the wrong problem. For example, a fish isn’t trying to escape the fishbowl, it’s aiming for a bigger fishbowl
- Purple ketchup and blue Pepsi were mistakes
- Innovation is gambling and much riskier, you win 50% of the time by gambling on red or on black
- Is the future knowable, it is not predictable, how do you know what will happen if you drop an egg off the roof… you predict it will break because you have a theory of gravity, science, you even know how fast it will fall
- Is there a science of innovation, what people will want, what innovations will succeed
- Need a theory that works in any market, any application, with significant risk reduction
- 90% of theories focus on the brain of the inventor, or you can look at evolution of systems, a sequence of steps, from phonograph to record to tape to disk to mp3
- evolution has order, it’s not purely random
- evolutionary forecasting needs – evolution is not random, set of natural laws
- evolution forecasting of the lawnmower – customer experience matters, forecasting is identify future problems and solve the problem, how do you ID the problem?
- How do you avoid wasting time by needing more oil, needing to sharpen blades, needing to try to start it 50 times, how to avoid me walking the machine over the lawn (robot mower)
- [more content in webinar but I had to leave]
Live note-taking of the November 9, 2016 webinar. Any errors are my own.
- Survey insights have been overshadowed in recent years, market research is struggling to redefine itself, there is an opportunity to combine big data and surveys
- Preferences are not always observable in big data, includes social data, wearable data
- Surveys can measure attitudes, preferences, and perceptions
- Problem is organizational – isolation, compartmentalization of market research and big data functions
- Started with a primary research survey about health and nutrition, one question is how often do you consume organic foods and beverages; also had block-group census data from American community survey five-year summary data with thousands of variables
- Fused survey data and block group data using location of survey respondent from their address and matched to block group data, brought in geo data for that block group
- Randomly split the data, built predictive model on training model, determined predictive accuracy using validation data (the hold-out data), 70% of data for model development, 30% for validation – independent objective model
- Created a lift curve, predictive model identified consumers of organic foods more than 2.5 times better than chance
- When predictive models bows out from random model, you have achieved success
- Which variables were most predictive, not that they’re correlated but they predict behaviour – 26 or older, higher income, higher education, less likely Hispanic; this may be known but now they have a model to predict where these people are
- Can map against actual geography and plot distances to stores
- High-tech truck research
- Used a choice modeling survey, design attributes and models to test, develop customer level preferences for features of the truck
- Cargo space, hitching method, back up cameras, power outlets, load capacity, price
- People chose preferred model from two choices, determined which people are price sensitive, or who value carrying capacity, biggest needs were price, capacity, and load
- How to target to these groups of people
- Fused in external data like previously, but now predicting based on choice modeling not based on survey attitudes, lift curve was again bowed to left, 1.8 times better than chance – occupation, education, income, and household size were the best predictors
- [these are generic results – rich people want organic food and trucks, but point taken on the method. If there is a product whose users are not obvious, then this method would be useful]
- Fusion can use primary and secondary data, also fuses technology like R dashboards and google maps, fuses survey and modeling, fuses consumer insights database marketing and big data analytics
- Use this to find customers whose preferences are unobserved, improve targeting of advertising and promotions, optimize retail location strategies, predict preferences and perceptions of consumers, collaboration of MR departments with big data groups would benefit both entities
- In UK and Spain, demographics are more granular, GPS tracking can be used in lesser developed countries
- Used R to query public data set, beauty of open-source code and data