The Global Source for Social Media Researchers

SMRA Blog 

  • 9 Jul 2019 11:53 AM | Kathy Doering (Administrator)

    Our friends from Talkwalker and Social Intelligence Lab came together to create an awesome infographic! Check it out below:

    AUTHOR: Meg Carpenter@Talkwalker

    Social data - raw insights collected from users’ social media activity - is key to helping you understand how to improve your social media performance, and maximize your digital marketing efforts. Every 'Like', every share, every click-through on your social posts means something, and the more you understand about such activity, the better you can focus your efforts on the processes which will ultimately drive your business' bottom line.

    But how are businesses actually using social intelligence, and what are the key trends to be aware of in the rising social data space?

    The teams from Social Intelligence Lab and Talkwalker recently partnered up to conduct a new report on "The State of Social Intelligence", in which they surveyed industry experts to get a better sense of how social data and analytics are being used to optimize performance.

    The report (which you can download here) includes some interesting trend insights, which may help to guide your own thinking and approach on social data usage - check out the infographic below for some of the key findings.  



  • 3 Jul 2019 12:19 PM | Kathy Doering (Administrator)

    Great new study from Social Media Examiner. Well worth the 5 minute read!

    Author: Elaine Fogel @elaine_fogel

    May 10, 2019

    Social media is one of the most popular marketing channels existing today. Yet, some business and organization leaders have questioned how effective it is in reaching their marketing objectives. Is it worth the financial and human resources investment?

    The newly released 2019 Social Media Marketing Industry Report from Social Media Examiner provides us with some insight on social media usage and behavior. The study surveyed more than 4,800 marketers with the goal of understanding how they’re using social media to grow and promote their businesses.

    Increased exposure and traffic are the top social media marketing benefits for the past five years. Increased exposure grew to 93% from 87% in 2018 and increased traffic improved to 87% from 78% last year demonstrating that positive results are possible even in a crowded social media landscape.

    The next two important increases:

    • Generated leads increased to 74% from 64%.
    • Improved sales rose to 72% from 53%.

    Even though Facebook and Instagram are the top two platforms used by marketers, it’s worth noting that Instagram grew from 66% to 73% and Twitter fell from 62% to 59%.

    Facebook is number one for both B2C and B2B marketers, however, when separating B2C and B2B responses, the number two spot for B2C is Instagram at 78% while B2B's second spot is LinkedIn at 80%.

    Overall, YouTube is still the number one video channel marketers prefer, with Facebook's native videos coming in second. When the study separated B2C and B2B responses, B2C marketers use more Instagram stories and Facebook native video while B2B marketers use more LinkedIn native video.

    Of the platforms marketers regularly use for social media ads, Facebook is way ahead of the other contenders. However, when separating B2C and B2B, the study shows that B2C marketers are more likely to use Facebook and Instagram ads while B2B marketers are using more LinkedIn ads.

    As with many studies of this nature, I always question the ROI (return on investment) for marketing activities. Is social media marketing worth the financial and human resources investment?

    Even though we see some ROI results in the first chart, “Benefits of social media marketing,” this next chart demonstrates how marketers need to uptheir measurement activities.

    In response to the study's question, “I am able to measure the return on investment (ROI) for my organic social media activities,” only 44% agreed they were able to measure their organic social activities. This is a challenge that has plagued marketers for years. 

    Social media marketing is a necessity in most marketing mixes today. Yet, the return on investment still eludes many of us. 



  • 28 Jun 2019 1:03 PM | Kathy Doering (Administrator)

    Here we go! Wonder if Social Media Research will indicate who the Democrat nominee will be....


    BY ANTHONY COSPITO 4 MINUTE READ

    Unlike any election before it, 2016 was a wake-up call for the role of social media in politics. This time around, voters and candidates alike have their eyes wide open to the promise and the peril these platforms can bring. That said, which Democratic candidates among the myriad running for election in 2020 are the stars of social?

    To find out, we at Moving Image & Content sorted the top six 2020 Democratic candidates for president by the unique donors and qualifications for debate, then ranked them by reach, growth, engagement, and activity online from April 25th to May 25th. Our results revealed the following contenders (in alphabetical order): former vice president Joe Biden, Pete Buttigieg (D-IN), Kamala Harris (D-CA), Amy Klobuchar (D-MN), Bernie Sanders (D-VT), and Elizabeth Warren (D-MA).

    While their approaches to social vary, all share a hyper-focus on Facebook—and with good reason. Yes, teens have left the platform in droves and fake news can still be found there, but 74% of U.S. adults continue to use the social network daily, according to Pew Research, making it an ideal proxy political platform for the 2020 nomination.

    Sanders is running a full-court press across Facebook. He’s currently leading in audience size with 5 million followers according to proprietary data by social analytics platform Rival IQ, and posting an average of 14 times a day on the platform. That’s three times more than any other candidate. But it’s not just the volume of posts that’s getting attention, it’s what he talks about that’s sparking a reaction. Surprisingly the loudest voice on the abortion ban isn’t Warren, Harris, or Klobuchar–it’s Sanders. His posts get more “angry” reactions than any other candidate (reactions that relate to the content of the post, not to Sanders himself) sparking waves of passionate conversation among family and friends. This is exactly what the Facebook algorithm favors and rewards.

    Also coming in strong is “Uncle Joe” Biden with his recent entry into the race. If the love Facebook users are showing for Biden’s announcement is any indication of where the polls are headed, Sanders better prepare for battle. Biden’s campaign announcement post got 15,000 “love” reactions–more than any of the other top six candidates. How voters interact with content is measured by engagement, including reactions, comments, and shares. This is arguably the most important metric for brands and politicians alike. While views of a post can be purchased, engagement cannot.

    Sanders leads in total engagement on Facebook at 3 million interactions, followed by Warren, Harris, Buttigieg, Biden, and Klobuchar. In terms of growth, however, the picture changes. When Biden entered the race, it fueled engagement growth of 2,890% compared to Warren (62%), Klobuchar (39%), and Sanders (12%). Looking at post comments, Biden leads growth at 1,700% compared to 32% from Elizabeth Warren and 8% from Sanders.

    In post shares, we see a similar story. Biden is up 8,000%, compared to Warren at 29% and Sanders at 14%. Biden currently has a 9-point lead over Sanders in share-of-voice across Facebook, Instagram, Twitter, and YouTube according to proprietary data from social listening platform Synthesio. Biden’s lead also shows an upward trend of 2.4% compared to Sander’s shrinking share-of-voice from 48% to 25% after Biden entered the race. Though Biden is clearly benefitting from the momentum, his challenge will be maintaining this level of growth. Buttigieg shows the largest loss in share-of-voice among the six at -6.48, which is a clear wake-up call for the candidate to change things up.

    Given the rising popularity of Instagram, candidates performing well here have a shot at winning over younger audiences. In terms of followers, Sanders leads at 3 million, followed by Harris at 1.7 million, Warren at 1.4 million, and Biden at 1.3 million. Sanders also leads in total engagement there, due in part to his strategic use of Instagram Stories content over the feed. This is a strategy both Warren and Harris share, helping them rank second and third in Instagram engagement respectively.

    Although Twitter reaches fewer voters than Facebook or Instagram, Trump has raised its profile as a platform for all things politics. Sanders leads here in followers, while Warren tweets the most and shows the strongest engagement overall. Sanders’ focus on social issues continues on Twitter. His tweet, “Abortion is a constitutional right” received the most engagement, outpacing top tweets from Warren and Harris about Attorney General Barr’s handling of the Mueller report. Biden’s announcement tweet ranked fourth in engagement, receiving over 30k retweets.

    Social performance metrics aside, how do the candidates fare in the friendliness factor? Examining positive and negative sentiment, Biden dominates both, indicating strong word of mouth and a healthy level of debate, followed closely by Warren and Sanders.

    As constituents, we need to hold our elected officials (and those running to become elected officials) to their promises, challenge proposed policies, and use social media to press them publicly on the issues that matter most. This empowers us to strengthen our democracy and leverage social media to fuel public discourse, not destroy it.

    Given the candidates’ current performance in these early days, a battle between Sanders and Biden appears to be brewing, with Warren keeping both on their toes.

    While the race continues to narrow, having a better understanding of how the candidates use social media in their quest for the nomination enables us to confidently consider our vote, phone in hand, eyes wide open.

    Anthony Cospito is the head of strategy at Moving Image & Content.


  • 26 Jun 2019 2:29 PM | Kathy Doering (Administrator)

    Interesting study by Stanford on School Shootings and Tweets from Politicians. 

    by Alex Shashkevich, Stanford University

    Analyzing the tweets of Republicans and DemocratsDora Demszky. Credit: Csenge Török

    New Stanford linguistics research has analyzed how Republicans and Democrats use different language when discussing mass shootings on social media and found that Republicans talk more about the shooter and Democrats focus more on the victims.

    Focusing on posts shared on the social media platform Twitter, the researchers found that Republicans tended to concentrate on breaking news reports and on event-specific facts in their tweets while Democrats centered on discussing potential policy changes, according to the new study, presented at a computational linguistics conference in June.

    "We live in a very polarized time," said the study's co-author Dan Jurafsky, professor of linguistics and of computer science. "Understanding what different groups of people say and why is the first step in determining how we can help bring people together. This research can also help us figure out how polarization spreads and how it changes over time."

    Researchers examined 4.4 million tweets posted in response to 21 different mass shooting events, including the Orlando nightclub shooting in 2016, to determine what words and emotions people with different political leanings expressed.

    They found that Republicans were more likely to express fear and disgust in their tweets than Democrats, who were more likely to communicate sadness and calls for action. Republicans were also 25 percent more likely than Democrats to write "terrorist" in tweets about the shootings in which the shooter was African American, Hispanic or Middle Eastern. Democrats were 25 percent more likely to use the same word when they tweeted about shootings in which the shooter was white.

    Studying tweets

    Researchers launched the study because they had three main questions: What is different about how Democrats and Republicans talk on Twitter? Could Republicans or Democrats be identified based on particular words they use in their tweets? How could these differences help understand the causes and consequences of social media polarization?

    To answer those questions, researchers used a method developed by Stanford economist Matthew Gentzkow together with Brown University economist Jesse Shapiro, who are co-authors on the new study, and economist Matt Taddy. The method determines the degree of polarization in speech, and it was used in previous research that examined the speech of members of Congress.

    The researchers applied the method and a language processing framework they created to a database of 4.4 million tweets about 21 mass shooting events that happened between 2015 and 2018. The researchers excluded retweets, and they determined whether a Twitter user was a Republican or Democrat by analyzing if they followed more Republican or Democratic politicians' accounts.

    Researchers chose to focus on responses to mass shootingsbecause "they are events with objective facts, the meanings of which people twist in different ways," said Dora Demszky, the lead author on the study and a Stanford linguistics graduate student. The interdisciplinary team of co-authors also includes James Zou, assistant professor of biomedical data science, linguistics graduate student Rob Voigt and electrical engineering graduate student Nikhil Garg.

    Researchers found that when people mentioned an earlier shooting as a way to contextualize the new shooting, Democrats were 2.7 times more likely than Republicans to mention a previous school shooting, most often the 2012 Sandy Hook Elementary School shooting. But Republicans were 2.5 times more likely to mention an event of mass violence that involved a perpetrator who was a person of color, which most often involved a mention of the Sept. 11 attacks.

    Researchers saw that the degree of polarization in the tweets increased over time in the hours and days following the events. For the three events where there was sufficient long-term data to draw conclusions, polarization plateaued usually after about three to four days, Demszky said.

    "Ideological polarization happens very fast," Demszky said. "As soon as an event like a mass shooting happens, people react very differently right away. This research gives a large-scale insight into how polarization works linguistically."

    Among other findings, researchers found that Democrats were more likely than Republicans to use phrases like "need to," "should," "have to" and "must" as part of their calls for political action.

    The research study also confirms previous research showing the relationship between people's beliefs, personalities and worldviews. The new study reveals that different emotions are expressed by people with different political leanings.

    Limitations and further research

    While some of the difference in speech patterns between Republicans and Democrats may be intuitive, the new study is one of the first to quantify polarization of language on social media in the hours and days after major events, Jurafsky and Demszky said.

    "In order to think about how we could fix the echo chambers that social media creates, we need data on how polarization happens," Demszky said.

    Further research is needed to understand the linguistic differences among Republicans and Democrats.

    One limitation of the new study is that researchers categorized each Twitter user they analyzed as either Republican or Democrat rather than locating them along an ideological spectrum.

    Demszky said she hopes that talking about language bias could be helpful in and of itself.

    "It's easy to not reflect on the words you use daily," Demszky said. "But I think it's a good step forward if people are just aware of their own biases."



  • 13 Jun 2019 9:49 AM | Kathy Doering (Administrator)

    SMRA is happy to bring some exciting research from one of our members:

    5th of June 2019 Michalis Michael Business Intelligence

    This is a short story about social intelligence (SI) and banks…

    listening247 – DigitalMR’s social intelligence solution – has a unique selling proposition: its high multilingual accuracy for brand, sentiment and topics annotation. Unfortunately, this unique selling proposition is also one of listening247’s biggest obstacles to scale. This trade-off between accuracy and scalability was of course not a surprise, it was a conscious decision made by a team of people who (as market researchers) have a tremendous respect for data accuracy; sometimes to their detriment. Thankfully, one day not too long ago, they realised that scalability does not have to be a trade-off for high accuracy.

    With listening247, a standard 3-week time period is usually needed to create new custom machine learning models each time new product categories and languages are to be analysed; new being the operative word. This is exactly how DigitalMR has been able to reach higher brand, sentiment and topics accuracy than competitors on the market. What the team of researchers realised was that once the custom machine learning setup (specific to a product category and language) exists, they could be on the same footing as every social media monitoring tool on scalability, but with a much higher accuracy!

    To put the theory to the test, one industry vertical was selected, and the necessary setup was created. That industry vertical was Banking.

    Why Banking?

    The decision was by no means easy, there were many variables to consider - perhaps too many; a strawman proposal was created and shared with the entire company and its advisors, and after a few weeks of back and forth, the banking sector was chosen.

    It would have perhaps been easier to choose a straight forward FMCG product category, or maybe retail, healthcare, automotive, or telecoms, but there are many good reasons why this vertical deserves focus. Those involved felt it’s time for banks to be enlightened as to the possibilities available to them in understanding their customers and competitors alike.

    Here are some factors that influenced the decision-making process in favour of banking:

    Banking is a particularly interesting sector as it offers both B2C and B2B services

    DigitalMR had produced SI banking reports in the past, in 2011 for the UK and in 2012 for the USA, and therefore already had a good understanding of the category and what would be involved in the semantic analysis (topics) setup in English.

    More recently DigitalMR had also produced a report on banks in Hong Kong, analysing social data in traditional Chinese.

    Within this vertical there are multiple sub-categories which exist as independent business units with P&L reports and their own budget to invest in business intelligence such as retail, corporate, wealth management, credit cards, insurance etc.

    A relevant opportunity with an academic institution was in sight, making the timing right. This was a R&D project about understanding how ESG – G for governance in particular – has an impact on a bank’s business performance. In case you are not familiar with the acronym ESG (a popular topic of conversation within the financial services sector) E stands for environmental and S stands for social.

    The legacy banks in this sector lack innovation and are increasingly being disrupted by challenger banks, blockchains and AI. If banks don’t already feel the urgency to make some drastic changes in the way they operate, they really should, and someone needs to help them get on the right path.

    The scope of the 1st report on Social Intelligence for Banks in 2019

    The process started by selecting 11 major and mostly multinational banks with the help of a high-profile industry advisor, to be used as keywords for data harvesting.

    Language: English

    Geography: Global

    Time Period: past 12 months

    Data sources: Twitter, blogs, boards / forums, news, reviews, videos

    Machine learning annotations: sentiment, topics, brands, and noise (irrelevant posts picked up due to homonyms)

    Deliverables: CSV with annotated data, Excel tables, drill-down and query dash boards, PowerPoint presentation

    As part of the R&D project with the academic institution, the daily valuations of each of the 11 banks were also retrieved from Yahoo Finance and Google Finance and taken into account.

    Volume of online posts justifies tracking bank performance online

    A total of 4.5 million English posts was harvested for the 11 banks globally. As you can see in the chart below, Twitter is by far the largest source of posts, followed by News – the only non-consumer source with mostly editorials published by journalists (or even the banks themselves).

    Source Distribution

    Looking at each bank individually, Twitter remains the largest source of posts for Deutsche Bank, HSBC, BNP Paribas, Santander and Credit Agricole, indicating that consumers do actually talk about their banks online, especially when they have complaints. On the other hand, the largest source of posts for banks such as Barclays, Société Générale, UniCredit and Intesa Sanpaolo is News, implying that either the customers of these banks have no complaints to share, or that they simply do not focus on engaging with them on social media.

    Source Distribution by Bank

    The first presentation & report highlights

    These findings were originally presented to a group of board directors of banks from various countries, who were taking part in the International Directors Banking Programme (IDBP) at INSEAD. Below are some of the report highlights.

    Deutsche Bank ranks first in buzz

    Deutsche Bank ranks first in terms of buzz (i.e. total volume of posts) with over 1.9 million posts coming from all sources. This represents 48% share of voice for Deutsche Bank, with HSBC and Barclays in 2nd and 3rd place respectively.

    Volume of Posts

    RBS has the lowest NSS™

    The Net Sentiment Score™ (NSS™) is a trademarked DigitalMR metric that combines all positive, negative, and neutral posts offering a composite metric that can be seen as the social equivalent of what NPS is to surveys. The Royal Bank of Scotland has the lowest NSS™ out of all 11 banks with a score of -3%, while HSBC leads the pack with a score of +9%. Interestingly enough, compared to other verticals or product categories a top NSS™ score of 9% is quite low.

    Net Sentiment Score - Banks

    Emotional Connection and ESG are hot topics

    When it comes to topics of online conversations and consumer sentiment in regards to those, financial events scored a NSS™ of -8%. ESG scored a +5%, while the best performing topic of emotional connection scored a +19%. Once again, ESG seems to be a very hot topic around banks and other corporates.

    Net Sentiment Score - Topics

    Deutsche Bank and RBS negative NSS™ for ESG

    This report can be quite granular in terms of topics and time periods. The below table drills down into the topic of ESG, showing the NSS™ for each bank (in relation to ESG) by quarter. Colour coding makes it very easy to pinpoint problem areas. In this case, Deutsche Bank and RBS are the two banks with the most quarters showing a negative NSS™.

    Net Sentiment Score - ESG

    During the event presentation it seemed as if the board level executives present had never seen anything similar in the past, taking in the results with a dose of scepticism, and of course asking several questions. Some wanted to drill-down and understand more about the findings from this data; those coming from the banks included in the project more so than others. The question is: will they be able to persuade the management of their banks to integrate social intelligence into all the other data streams they currently use?

    What makes this report credible is that we know its sentiment and topic accuracy is over 75%. This is not just a claim, a human can verify it by extracting a random sample of 100 posts, reading through them, and checking their agreement with the machine learning algorithm annotations (brand, sentiment, and topics). There should be agreement with any of the above annotations for at least 75 posts.

    The next story

    This is only the first of a series of short stories on this banking report. By the time the next one is published, our machine learning models will have improved themselves to reach accuracies over 80%.

    If you are wondering what other ways there are to create value for a bank from a social intelligence report as such, stay tuned! In the next article you can expect to find out how news about governance impact the valuation of one of the 11 banks in this report. For anyone too impatient to wait another couple of weeks, feel free to reach out to me on Twitter @DigitalMR_CEO or email me at mmichael@digital-mr.com. Talk soon!



  • 9 Jun 2019 5:13 PM | Kathy Doering (Administrator)


    In today’s advanced digital marketplace customers are no longer stuck with only an automated phone system to respond to their customer service needs. There are a wide variety of service channels to choose from that cater to different customers - self-service knowledge bases, chatbots, and social media messaging services.

    And it seems that companies are seeing the value of social media customer service. In their “2016 State of Social Business” publication, Altimeter analysts Ed Terpening and Aubrey Littleton report that among 523 respondents in big companies in the US and Europe “social customer service” is now the top external objective for social business functions, just ahead of “relationship building” – which also focused on current customers, not customer acquisition.

    For customers, interacting with brands via Twitter, Facebook, Instagram, Messenger, etc. is often faster and easier, and many customers believe they receive better care via these channels than they receive via phone and email. According to a new report from Conversocial, 54% of customers prefer customer service via social media and SMS. Because of this, Twitter and Facebook have rolled out several tweaks and enhancements to their platforms to make them even more viable replacements for traditional contact channels.

    So what is Social Media Customer Service?

    Social customer service is the practice of providing consumer support through social media channels to quickly answer questions. 69% of customers believe fast resolution of the problem is vital to good service, making social consumer support invaluable. While Facebook and Twitter have proven to be vital platforms for marketing, they are also important channels through which consumers solicit and receive customer service. According to the Q2 2016 Sprout Social Index, 90% of surveyed consumers have used social media in some way to communicate with a brand. And, over 1/3 (34.5%) said they preferred social media to traditional channels like phone and email.

    How can you improve your Social Media Customer Service?

    1) Be where your customers are.

    Where should you focus your time and efforts? Facebook and Twitter will be the primary focus for most companies, but other brands may find that their customers also frequent Google+, LinkedIn, Pinterest, Instagram, or other social sites.

    To find out where your audience is, search for mentions of your brand within popular social sites. If you find that your audience isn't yet talking about your brand online, look for ways to include yourself in conversations relevant to your industry. The way, for an employee to be welcomed into social conversations is to add something of value.

    2) Listen to what your customers have to say.

    Many marketers are already familiar with social media monitoring tools that automate the process of searching for mentions of a brand name, but listening is equally important from a customer service perspective. Research from the Institute of Customer Service reveals 1 in every 3 customers turns to social media to seek advice or communicate with a business.

    Depending on how much volume your brand's social media pages generate, it's important to collect and analyze activity so that you understand the issues being raised over social media. This information allows you to determine:

    How many comments reflect a poor customer experience, in person or online?

    How many comments provide feedback, positive or negative?

    How many brand mentions require, or would benefit from, a response?

    What time of day are your customers most active on social media?

    The answers to these questions will help you define priority content, make decisions about self-service options, and determine whether you'll be able to handle the majority of issues directly through the social media channels or directing users to other lines of support.

    3) The speed of your response is critical.

    Studies have shown that most customers want a response over social media within the same day. The Northridge Group reported that 42% of consumers expect a response to their customer service inquiry within the hour. Because tweets and timeline posts can be submitted overnight, this presents a challenge by driving your response time from just a few hours to 10-20 hours later.

    As a best practice, always respond with immediacy—or with the promise of. A good strategy may be to set up an automated response letting customers know you received their message and will respond the next business day. At least then they know their message was received and you are working on a resolution.

    4) The success of your social care efforts will depend on the quality of care you provide.

    Agent responses must be timely, accurate, sensitive, brief, and friendly…which is a lot to ask. Agents must be able to read into a customer's emotional state and determine an appropriate response. That may involve a message conveying friendliness and willingness to help or possibly sending a more formal statement of empathy or apology to address an issue.

    So what can you do when you receive negative feedback? This is an opportunity to rectify your brand's image and, more important, your relationship with the customer. The customer must feel like they've been heard and that you're willing to do what it takes to make them happy. Research for Hug Your Haters (conducted by Edison Research) studies found that customers are more likely to advocate on behalf of brands who answer their complaints.

    When a company answers a customer complaint via email, it increases advocacy by an average of 8%.

    When a company answers a customer complaint via phone, it increases advocacy by an average of 10%

    When a company answers a customer complaint via social media, in increases advocacy by an average of 20%.

    Regular monitoring of your company's social media pages combined with savvy use of the sites can elevate your customer service efforts from acceptable to exceptional. The better your social care, the more social traffic you can expect, and ultimately more loyal customers.


  • 30 May 2019 6:53 PM | Kathy Doering (Administrator)

    The medical industry is gaining insights into all types diseases and medical conditions by using social listening insights. I really like the time line for the Diabetes study too. This is a great example of how social listening is done correctly. 



    Reference

    Lee C. Gaining insight into the patient’s experience by harnessing the power of social listening and FDA archival data. Presented at: FDA Grand Rounds; May 9, 2019; Washington, DC.

    This article originally appeared on Medical Bag

    When Christine Lee, PharmD, PhD, a health scientist with the US Food and Drug Administration’s (FDA’s) Center for Drug Evaluation and Research, wanted information about caring for her newborn baby beyond what was captured in parenting books, she turned to social media. She joined a forum for new mothers on Facebook to ask the crowd for advice.

    Soon after, she wondered how social media platforms such as Facebook, Twitter, Instagram, and Reddit can be mined for data that could help the FDA’s pharmacovigilance efforts. “I thought, ‘Could I apply the same qualitative research methods traditionally used for focus groups and cognitive interview data to research involving unstructured narrative data in social media?’ ” explained Dr Lee during her grand rounds presentation, “Structuring Unstructured Data: Using New Data Sources to Understand the Needs of Underserved Populations,” presented May 9, 2019, in Washington, DC.

    How the FDA Uses Patient Experience Data

    The agency’s regulatory mission relies on sourcing new data and methodologies to increase its understanding of patients’ perspectives. The FDA uses patient experience data to inform:

    • Clinical trial design

    •  Trial end point development and selection

    • Regulatory issues, including benefit-risk assessments

    To meet patients’ needs, the FDA strives to engage patient stakeholders throughout the life cycle of a medical product. Social media platforms provide a method for capturing meaningful, unfiltered patient insights, according to Dr Lee. It can give the FDA a more comprehensive picture of how medical products function beyond controlled, randomized clinical trials.

    The FDA has 2 objectives for engaging patients:

    1. Support the FDA’s and the Center for Drug Evaluation and Research’s goals of understanding the patient’s voice, including their perspectives on conditions and treatments.
    2. Refine qualitative research methods to explore tapping unstructured data with high repeatability.

    Related Articles

    Social Media Monitoring: Gaining Insight Into Diabetes Therapies

    The US Department of Health and Human Services’ Office of Minority Health funded a pilot study to analyze social media posts for useful data about minority patients with diabetes and their treatment.

    During the last 2 years, the agency used data-mining software to collect, monitor, and analyze more than 100,000 conversations on Twitter. The effort targeted a variety of keywords, including diabetes, glucose, diabetic, and blood sugar. Demographic filters were added to narrow the query. To separate signal from noise, tweets that were classified as cheeky, advertisements, or spam were removed, leaving about 73,000 tweets for analysis. The remaining tweets were scrubbed and formatted, and then natural language processing and machine-based learning were used to help researchers identify trends. This process was also used to uncover trends in FDA Advisory Committee data, and then the 2 data sets were examined.

    Improving Diabetes Education Among Minorities

    Researchers sought to discern what government organizations, patient advocacy organizations, and pharmaceutical companies share on social media, and whether this information is specific to racial or ethnic minorities. The data revealed several content themes:

    • Awareness: 39.9% of tweets analyzed

    • Diabetes management: 22.4%

    • Risks associated with diabetes: 13.3%

    • Diabetes prevention: 7.3%

    • Other: 9.1%

    “We found that little to no discussions on Twitter were specific to minority groups,” Dr Lee said. “Our research suggests that there is an opportunity to improve outreach to minority groups that is specific to their unique health needs.”

    Researchers also concluded that there are gaps in their understanding of minority groups’ perceptions of the risks associated with FDA-regulated prescription drugs, their unique health needs, and level of health literacy.

    To gain further insight, researchers mined a sample of Facebook posts from January 2017 to June 2017 that mentioned diabetes. After compiling a list of the top 50 most-liked posts for each month, the content was analyzed to elicit meaning from the collected posts. An extensive codebook was developed to guide this process. The data highlighted the importance of providing a patient-centered approach to care and individualizing care, according to researchers. More specifically, the data demonstrated the importance of:

    • Educating patients on how diet and exercise can help them manage their diabetes.

    • Improving patients’ awareness about comorbid conditions and teasing out related symptoms.

    • Improving awareness about how the link between diabetes and the household environment can increase the risk of developing diabetes.

    • Providing patient support and encouraging community engagement to promote healthier environments.

    Although Dr Lee was encouraged by the enhanced patient insight she and her team were able to gain from social media posts, she cautions that social media data must be combined with other sources of data, including FDA archives, patient-focused drug development data, public docket comments, advisory council transcripts, focus groups, and listening sessions. According to Dr Lee, the findings of this study suggest that new data sources can increase the understanding of the patient perspective, particularly of vulnerable populations, and can increase confidence in the data that the FDA traditionally collects.

    “Social media is a useful data source to gather relevant patient perspectives on barriers and may provide information from populations who may not utilize FDA sources,” Dr Lee concluded.

    The FDA also conducted a social media listening pilot project on Reddit to gather data regarding opioid use. The results of this study are under manuscript review and cannot yet be shared publicly.


  • 20 May 2019 4:42 PM | Kathy Doering (Administrator)

    Ever heard the term Big Brother is watching? What you are about to read regarding your smart phone and specific apps may alarm you.

    We use apps constantly without thinking twice about what information or knowledge they may be gathering about us. But do you know that some smartphone apps are used to track our every move? Thanks to tiny pieces of code that millions of developers use to make their lives easier, an array of companies gets free access to data they can employ to understand your habits.

    When we browse the web through Google Chrome, for example, vast arrays of companies follow us. That’s why the ads that appear on the right side of your browser are usually related to searches you have recently conducted.

    On your smartphone, tracking is generally performed through the use of a “software development kit” or SDK—a set of tools that help developers debug their code or hook into useful services. But other SDKs help advertisers and marketing companies peer into your private life. Take the iHeartRadio app for example: Last fall, Medium reported that it contained code from Cuebiq’s SDK, which would permit user data to be sold for the purposes of ad tracking.

    Apple just recently launched a “privacy matters” campaign, which is ironic because it doesn’t protect users from trackers embedded in apps that are distributed through the iOS App Store. SDKs also allow Facebook and Google to track users beyond their desktop web browsers and automatically collect information like when you installed the app, each time you opened it, and what you purchased.

    Tracking in SDKs is clearly part of the modern App Store era and there are tons of companies you’ve never heard of invisibly tracking your habits in apps you use every day. Networks like Vungle, Apps Flyer, and Applovin all call themselves “advertising and analytics” platforms. They help developers monetize their apps, and all of them track data to sell to other partners behind the scenes as well.

    In the past, Apple has moved to make it more difficult to identify you by blocking access to unique identifiers and your phone number, but it’s still trivial to correlate an identity via your IP address, the name of a Wi-Fi network, or just matching together the bread crumbs of data they grab about you. Android allows even broader access to identifiers—not surprising, given that it’s built by a company that relies on advertising to make money.

    There’s frustratingly little we can do to combat SDK tracking without intervention from Apple and Google. They should provide operating system controls that show the parties harvesting data inside the apps on our devices or should require third parties to reveal this information. A good example of this in practice can be found in the Guardianapp, which allows users to disable tracking on a per-SDK basis in its settings. Requiring this should be standard for all developers.

    Let’s look at another example…more intrusive, but surprisingly popular among women.

    Ovia, a pregnancy-tracking app, allows users to record private and intimate details of their entire pregnancy journey including bodily functions, sex drive, medications, mood, ovulation cycle, and more.

    Ovia pitches its app to companies as a health-care aid for women to better understand their bodies during a mystifying phase of life. In marketing materials, it says women who have tracked themselves with Ovia showed a 30% reduction in premature births, a 30% increase in natural conception and a higher rate of identifying the signs of postpartum depression. Women wanting to get pregnant are told they can rely on Ovia’s “fertility algorithms,” which analyze their menstrual data and suggest good times to try to conceive, potentially saving money on infertility treatments. “An average of 33 hours of productivity are lost for every round of treatment,” an Ovia marketing document says.

    Employers who pay the apps’ developer, Ovia Health, can offer their workers a special version of the apps that relays their health data — in an anonymous form — to an internal employer website accessible by human resources personnel. The companies offer it alongside other health benefits and incentivize workers to input as much about their bodies as they can, saying the data can help the companies minimize health-care spending, discover medical problems and better plan for the months ahead.

    But some health and privacy advocates say this new generation of “menstrual surveillance” tools is pushing the limits of what women will share about one of the most sensitive moments of their lives. The apps, they say, are designed to benefit the employers and insurers and experts worry that companies could use the data to bump up the cost or scale back the coverage of health-care benefits, or that women’s intimate information could be exposed in data breaches or security risks.

    “The real benefit of self-tracking is always to the company,” says Karen Levy, a Cornell University assistant professor who has researched family and workplace monitoring. “People are being asked to do this at a time when they’re incredibly vulnerable and may not have any sense where that data is being passed.”

    An Ovia spokeswoman said the company does not sell aggregate data for advertising purposes. But women who use Ovia must consent to its 6,000-word “terms of use,” which grant the company a “royalty-free, perpetual, and irrevocable license, throughout the universe” to “utilize and exploit” their de-identified personal information for scientific research and “external and internal marketing purposes.” Ovia may also “sell, lease or lend aggregated Personal Information to third parties,” the document adds.

    For employers who fund workers’ health insurance, pregnancy can be one of the biggest and most unpredictable health-care expenses. In 2014, AOL chief executive Tim Armstrong defended the company’s cuts to retirement benefits by blaming the high medical expenses that arose from two employees giving birth to “distressed babies.”

    “The fact that women’s pregnancies are being tracked that closely by employers is very disturbing,” said Deborah C. Peel, a psychiatrist and founder of the Texas nonprofit Patient Privacy Rights. “There’s so much discrimination against mothers and families in the workplace, and they can’t trust their employer to have their best interests at heart.”

    Federal law forbids companies from discriminating against pregnant women and mandates that pregnancy-related health-care expenses be covered in the same way as other medical conditions. Ovia said the data helps employers provide “better benefits, health coverage and support.” Pregnant women can log details of their sleep, diet, mood and weight, while women who are trying to conceive can record when they had sex, how they’re feeling and the look and color of their cervical fluid.

    After birth, the app asks for the baby’s name, sex and weight; who performed the delivery and where; the birth type, such as vaginal or an unplanned C-section; how long labor lasted; whether it included an epidural; and the details of any complications, such as whether there was a breech or postpartum hemorrhage.

    The app also allows women to report whether they had a miscarriage or pregnancy loss, including the date and “type of loss,” such as whether the baby was stillborn. “After reporting a miscarriage, you will have the option to both reset your account and, when you’re ready, to start a new pregnancy,” the app says. “We’re their companion throughout this process and want to … provide them with support throughout their entire journey,” Ovia spokeswoman Sarah Coppersmith said.

    Much of this information is viewable only by the worker. But the company can access a vast range of aggregated data about its employees, including their average age, number of children and current trimester; the average time it took them to get pregnant; the percentage who had high-risk pregnancies, conceived after a stretch of infertility, had C-sections or gave birth prematurely; and the new moms’ return-to-work timing.

    Ovia data is viewable by the company, their insurers and, in the case of Activision Blizzard and other self-insured companies, the third-party administrators that process women’s medical claims. Ezzard, the benefits executive at Activision Blizzard, said offering pregnancy programs such as Ovia helps the company stand out in a competitive industry and keep skilled women in the workforce coming back. The company employs roughly 5,000 artists, developers and other workers in the United States. “I want them to have a healthy baby because it’s great for our business experience,” Ezzard said. “Rather than having a baby who’s in the neonatal ICU, where she’s not able to focus much on work.”

    “As a clinician researcher, I can see the benefit of analyzing large data sets,” said Paula M. Castaño, an obstetrician-gynecologist and associate professor at Columbia University who has studied menstrual-tracking apps. But a lot of the Ovia data given to employers, she said, raises concerns “with their lack of general clinical applicability and focus on variables that affect time out of work and insurance utilization.”

    Ovia says its “fertility algorithms,” which analyze a woman’s data and suggest when she would have the best chance of getting pregnant, have helped 5 million women conceive. But the claim is impossible to prove: Research into similar promises from other apps has suggested there were other possible explanations, including the fact that the women were motivated enough to use a period-tracking app in the first place.


  • 7 May 2019 7:50 PM | Kathy Doering (Administrator)

    Finally after years of criticism, Facebook announced that it would stop allowing advertisers in key categories to show their messages only to people of a certain race, gender or age group. The company said that anyone advertising housing, jobs or credit — three areas where federal law prohibits discrimination in ads — would no longer have the option of explicitly targeting people on the basis of those characteristics.

    Facebook has been accused of allowing advertisers to unlawfully discriminate against minorities, women, and the elderly by using the platform’s ad-targeting technology. The settlement resolves five separate cases that had been brought against Facebook over discriminatory advertising since 2016, following a ProPublica investigation that revealed Facebook let advertisers choose to hide their ads from blacks, Hispanics, or people of other “ethnic affinities.” Lawsuits soon followed. The most recent case was an EEOC complaint by the American Civil Liberties Union in September, alleging that Facebook allowed job ads to discriminate against women.

    This is significant because Facebook’s massive revenue primarily comes from ads, which are so lucrative because of their microtargeting capabilities. In 2017, according to its annual earnings report, the company made $39.94 billion on ads alone. Its total revenue for that year was $40.65 billion, meaning ads accounted for roughly 98% of revenue.

    But when a company or advertiser shows an ad only to certain people, it excludes a protected class of workers. And that’s illegal under federal law. “It is a game-changer,” says Lisa Rice, the executive vice president of the National Fair Housing Alliance, whose lawsuit against Facebook was among those settled Tuesday. “The settlement positions Facebook to be a pacesetter and a leader on civil rights issues in the tech field.”

    “We think this settlement is historic and will go a long way toward making sure that these types of discriminatory practices can’t happen,” Sheryl Sandberg, the company’s COO, said in an interview.

    As part of the agreement, Facebook will build a designated portal for advertisers to create housing, employment, and credit ads, which will not allow targeting users by age, gender, zip code, or other categories covered by anti-discrimination laws. Microtargeting options that appear to relate to these protected categories will be off-limits as well. Any advertiser that wants to run an ad on Facebook will be required to indicate if their ad is related to one of these three things. According to The Washington Post, Facebook has said it will make these changes by the end of the year.

    “Housing, employment, and credit ads are crucial to helping people buy new homes, start great careers, and gain access to credit. They should never be used to exclude or harm people,” Sandberg wrote in a post announcing the settlement. “Getting this right is deeply important to me and all of us at Facebook because inclusivity is a core value for our company.”

    Advertisers that deliberately and repeatedly avoid the new portal when placing ads in the three regulated areas will probably face consequences, though the company said it had yet to determine those.

    Pauline Kim, a professor of employment law at Washington University in St. Louis, praised the changes but cautioned against overstating their significance. “Taking the explicit ability to discriminate off the table is an important first step,” Professor Kim said. “But I don’t think it solves the problem of the potential for biased serving of ads.” Ms. Kim said, for example, that an employer could place an ad that it intended to show to both men and women, but over time, Facebook’s algorithms could begin to show the ad primarily to men if it determined that men were much likelier to click on the ad. “It’s within the realm of possibility depending on how the algorithm is constructed,” Professor Kim said. “You could end up serving ads, inadvertently, to biased audiences.”

    Sandberg acknowledged the limits of the policy changes and said Facebook had committed to working with the other parties to find additional ways to root out discrimination. The parties will discuss progress on that front every six months for three years after the changes are rolled out.

    “In addition to being a historic settlement of five separate lawsuits that will change practices on Facebook and other platforms, it’s also notable that we agreed to continue to study the algorithmic effect of ads with Facebook,” said Anthony Romero, executive director of the A.C.L.U.

    Facebook is also removing thousands of so-called interest segments, including some that advertisers could use to reach people by characteristics like gender or age, for ads in the three regulated areas. For example, to show how advertisers could keep certain groups from seeing housing ads on Facebook, the National Fair Housing Alliance once created a fictitious ad that excluded groups like “corporate moms” and “stay at home” mothers.

    Those segments would no longer be available for targeting by housing, employment and credit ads once Facebook carries out the proposed changes.

    Despite some of the expected pushback, civil rights advocates are applauding and they are confident Facebook will follow through. The company has agreed to twice-annual meetings with the groups, as well as ongoing trainings with outside experts on these issues. Facebook has agreed to let the NFHA, the ACLU, and others conduct independent testing of its ad sites to make sure Facebook does what it says it will.

    “If any advertiser was trying to skirt or circumvent the system, we have methods for finding that out and we’ll be able to bring that to the attention of Facebook,” says Rice of the National Fair Housing Alliance.


  • 25 Apr 2019 5:09 PM | Kathy Doering (Administrator)

    By: Jim Matorin, SMARTKEING

    Proviso: Last year I harped on the value of marketers understanding their target audience either via conventional demographic categorization or digitally mining for brand advocates to implement influence marketing movements. I will also go on record: I stated that marketers are over processing when it comes to consumer targeting.  Time flies, summer is just around the corner. I still persist marketers will continue to over process when it comes to consumer targeting. Thanks to the advent of 5G (the next generation of mobile broadband) and the utilization of A.I., we will experience consumer targeting on steroids – psychographic profiling.

    A recent survey conducted by eMarketer Retail indicated a small percentage, only 5% of retailers and CPG enterprises are leveraging data in their decision making. Ironic given the survey respondents ranked the importance for better (89%) and faster (79%) insights into customer needs and expectations. Thanks to the expansion of data captured at the numerous consumer technology touch points – transactional purchasing behavior/history, apps, geo-location, etc., smart marketers are in stronger position to make data driven decisions based on enhanced, more effective segmentation. The new wave of segmentation will reflect attitudinal information (e.g., personal interests, life styles, occasion, etc.) layered with traditional demographic buckets (e.g., age, income, education, etc.).   

    I envision the advent of 5G, the next generation of mobile broadband, combined with the utilization of A.I. will enhance consumer segmentation targeting. 5G connections will deliver greater data capacity (instant computing power) and speed than previous generations – 1,000 the data rate of 4G. Consequently, marketers will be able to utilize psychographic profiling thanks to their ability to crunch data in nanoseconds to identify and engage with their target audience via relevant, real time marketing communications.

    My company specializes in the food industry. Detailed below are some examples of future food industry psychographic profiling buckets:

    • Lifestyles (e.g., “on-the-go”) – Snacking behavior, convenience foods preferences, beverage consumption.
    • Situational Eating Behavior – Location and time of day consumers buy food.
    • Health vs. Indulgence – Special diet needs, labeling knowledge, over the top eating experiences.
    • Social Values – Make the world a better place – sustainability, responsible biodiversity sourcing.

    Does psychographic profiling sound complicated? To a degree, except for those companies that invest in the resources, both technology and trained human capital (analysts). They will eventually develop an integrated segmentation methodology, the combination of key consumer profiling factors that drive their sales. However, given what I have experienced throughout the years, a majority of marketers will over process. Over processing is in their DNA. They will end up compiling mountains of data, conduct endless meetings, even go to length to hire outside resources (e.g., agencies) to help find the magic bullet.

    Bottomline: There is no magic bullet. Companies have to seek the marketing formula that works best for them. Execution is a great starting point.






Copyright 2017, Social Media Research Association. All rights reserved

Powered by Wild Apricot Membership Software