Up close and personal with consumers using nonconscious measurement and text analysis #IIeX
Live note-taking at #IIeX in Atlanta. Any errors or bad jokes are my own.
Chameleon communities: immersive learning takes many forms with Stephen THompons and Kendall Nash
- Communities are iterative sharing, exchange of ideas, genuine dialogue, relationship building, sense of accountability
- Chameleons can change colour, to communicate, to be what they need to be; communities should be what we need them to be
- COntinuous insight communities – reach out to them an any time, on-demand insights
- Longer term communities 3 to 4 weeks
- SHort term pop-up communities – bulletin boards sometimes, less than 3 weeks, part of a wider research program
- Use for testing, online diaries, advisory Angels, shop alongside, co-creation, feedback, and it’s better if they’re layered
- Technology accommodates learning
- Build an online environment o suit – easy to use, attractive, and engaging tools make research fun
- Great for discussions of TV shows and what is and isn’t funny
- Feel licensed to right size them and make the suit your needs, if it’s not perfect then make it right
What nonconscious measurement says about America as a brand by Elissa Moses
- America is a brand [I assume you mean USA because America includes Canada, Mexico, and Brazil]
- When brands erode, they become vulnerable
- Brands give you shortcuts re expectations and whether something will be a good fit
- Implicit research was great to study racial prejudices because they either don’t want to tell you, or they’re conflicted or they really just don’t know
- Brands can change – Disney, apple, VW, BP all used to be high esteem and fell from grace, vulnerability always lurks, there can be a scandal tomorrow
- Used a device agnostic tool, mobile friendly [Did you hear that MOBILE FRIENDLY! 🙂 ]
- Calibrated on individual baselines [thus repeated measures, within subjects design]
- Sweden, USA, MExico, France have high scores on great place to live
- but if you ask if their country will ever be as great again, then negatives pop up
- There is huge ambivalence about where the country is headed and that people worry about
- 50% of Americans say yes to beginning in immigrants [ of course, even if I take longer to answer a question, it could be not that am unsure, but that I want to make sure I truly believe my answer. I know I pause a lot when I answer questions because that’s how introverts do it]
- Tools is good for busting cliches
- People in America just aren’t sure about their country any more
- People don’t believe American standards for justice, upholds the constitution, respects separation of court and state, gives equal to rights to all
- People DO have pride in America so we can get back on track
Return on customer investment: linking customer insights to revenue growth by Manila Austin
- Are companies customer-centric? 80% of companies think they are but only 8% of customers think so
- Returns on Assets is declining and going flat, old models are no longer working
- Companies spend 14$ on advertising for every 1$ spend on understanding consumers
- Most R&D efforts go towards sustaining existing products
- When customers are more likely to say a company “Gets them” their revenues are more likely to increase in comparison to other similar companies
- Raising your customer quotient will increase your ROA, a .5 point increase is worth millions in revenue and net income
- Employees and consumers see things differently, employees thinks their customer experience is far better than consumers
- Consumers want openness, relevant, loyalty; employees want openness, empathy, to be closer to end users
- Diagnosed the customer into the organization
Social Insights: The next generation by Rob Key
- Language is really complicated
- Rules based solutions are complicated, one wrong word and everything gets messed up
- Precision – do they match gold standard of humans, 80% of time humans can tell what it means – 3 independent humans [I remember when no one cared about validation 🙂 ]
- Relevance – does a Boolean query do the trick? Maybe but it’s not nearly enough; words take on different meaning in different domains, small is good for smartphones but not for hotel rooms; faded jeans are good but not faded interiors; every industry has its own lexicon
- Need precision and relevance and recall to achieve quality
- People don’t say “I trust this brand” or “I highly recommend this brand” THey say things like “I give this to my baby” or “I’m the hell outta here”
- Language must be customized to industry
- Emotions come in many forms which many words [seems like everyone uses plutchik’s wheel of emotion 🙂 ]
- Can you isolate spam the food from spam the email hell
- Unify the data with call Center data, survey data, unification f the voice of the customer
- Social data is not quantitative and meaningful, mainstreamed into large organizations
- Clean data does indeed create valid results [as does clean survey data and clean focus group data]
Harnessing text for human insights #IIeX
Live note-taking at #IIeX in Atlanta. Any errors or bad jokes are my own.
Chaired by Seth Grimes
Automated text coding: humans and machines learning together by Stu Shulman
- It is a 2500 year old problem, Plato argued it would be frustrating and it still is.
- Coders are expensive, it’s difficult at scale, some models are easier to validation than others, don’t replace humans, no one right way to do it, validation of humans and machines is essential
- Want to efficiently code, annotate coding with shared memoirs, manage coding permissions, have unlimited collaborators, easily measure inter-rather reliability, adjudicate validity decisions
- Wanted to take the mouse out of the process, so items load efficiently for coding
- Computer science and HSF influence measure everything
- Measure how fast each annotator works, measure interacter reliability, reliability can change drastically by topic
- Adjudication – sometimes it’s clear when an error has been made, allows you to create a gold standard training set, and give feedback to coders; can identify which coders are weak at even the simplest task, there is human aptitude and not everyone has it, there is a distribution of competencies
- 25% of codes are wrong so you need to train machines to trust the people who do a better job at coding
- Pillars of text analytics – search, filtering, deduplication and clustering and works well with surveys as well, human coding or labelling or tagging which is where most of their work goes, machine learning – this gives a high quality training set
- If humans can’t do the labelling, then the machines can’t either
- Always good to keep humans in the loop
- Word sense dis ambiguities – relevant – is bridge a game or a road, it smoking a cigarette or being awesome
Automated classification interesting, at scale and depth by Ian McCarty
- Active data collection is specific and granular, as well as standardized; but it’s slow and difficult to scale, there is uncertainty, may be observer bias via social desirability, demand characteristics, Hawthorne effect [EVERY method has strength and weaknesses]
- Declared vs demonstrated interests – you can give 5 stars to a great movie and then watch Paul Blart Mall Cop 5 times a 6 months [Paul Blart is a great movie! Loved it 🙂 ]
- They replicate the experience of a specific URL to generate more specific data
- Closed network use case – examined search queries from members to recruit them into studies, segmentation was manual and company needed to automate and scale; lowered per person costs and increased accuracy, found more panelists in more specific clusters, normalized surveys if declared behaviors conflicted with demonstrated behaviors
- Open network use case: home improvement brand needed a modern shared meaning with customers, wanted to automate a manual process; distinguished brand follower end compared to competitive followers, identified where brand values and consumer values aligned, delivered map for future content creation and path to audience connection
Text analytics or social media insights by Michalis Michael
- Next gen research is here now, listening, asking questions, tracking behavior, insights experts
- Revenues don’t reflect expectations, yet.
- We’re not doing a great job of integrating insights yet, social media listening analytics is not completely integrated in our industry yet
- Homonyms are major noise, eliminating them needs humans and machines
- Machine learning is language agnostic, create a taxonomy with it, a dictionary of the product category using the words that people use in social media not marketing words
- It is possible to have 80% agreement with text analytics and the human [I believe this when the language is reasonably simple and known]
- Becks means beer and David beckham but you need training algorithms to do this, Beck Hanson is a singer, you need hundreds of clarifications to identify the exact Becks that is beer
- Beer is related to appearance and occasions, break down occasions into in home or out of home, then at a BBQ or club
- What do you say about a beer when they do a commercial that has nothing to do with the beer
- English has s a lot of sarcasm, more than a lot of other languages [yeah right, sure, I believe you]
- Break down sentiment into emotions – anger, desire, disgust, hate, joy, love, sadness – can benchmark brands in these categories as well
- Can benchmark NPS with social media
- Brand tracking questions can be matched to topics in a social media taxonomy, and there can be even more in the social media version than the survey version
The LoveStats Blog 