Product and Service Organizations are increasingly showing keen interest in sentiment analysis as it can make or break their reputation overnight. Understanding the polarity/sentiment of every users post and addressing them on time is becoming critical. Considering the fact that thousands of posts are being generated every day on a single topic, it is nearly impossible to analyze them manually.

Sentiment analysis algorithms are not accurate and cannot be 100% accurate. But it can certainly provide the required warning signals to change the (product) strategy if required at the right time. This can also help the organization save time by narrowing down the thousands of tweets to a few hundred tweets to be analyzed manually.

Once the need for sentiment analysis has been identified, the organization starts to scout for COTS products and find there are way too many. This article has a list of 40+ parameters under six key buckets that the organization can use to shortlist the right tool for them.

In the below table, you can find the six buckets mentioned above along with the parameters and a short description about each.

Feel free to comment on the above parameters. If you have any addtional parameters that you feel are relevant, do let me know and I will be happy to include them (with due credits to you).

INTRODUCTION

Everyone understands the value of grouping users. There are various ways to classify users but primarily they fall into one of the three levels defined below:

LEVEL 1: PRIMITIVE

This is the easiest and the most common way of grouping users. This Level is synonymous to a company in US trying to read books about Japan to understand their culture. The distance between the user and the organization is high. Beyond a certain point, this level fails to connect with the users.

Organizations can only take generic decisions based on the typical characteristics of this user group. Example, if an organization is developing a portal and it finds that 75% of its users are female and in the teenage group, they might target pink for the portal theme. But if this group happens to like rock music and partying, the organizations decision to go with pink could be a road to disaster.

LEVEL 2: PASSIVE

Levels-2 is all about studying the user passively but closely to understand their presence (where they are active), their behavior (what they do online) and their tastes (what they like). This is synonymous to a company in US sending someone to Japan to collect information. Inclusion of Level 1 grouping to this level would be an added advantage. A good starting point would be to analyze data captured from social media networks like Facebook, Twitter, etc.

LEVEL 3: ACTIVE

In this level, hand picked influencers are analyzed in depth from both an internal & external networks perspective. Right content (information, samples) is then pushed to this small group of handpicked few to create the impact. This levels termed “Active” because on identification of influencers, some amount of handshake/contact/personal touch happens with the influencer (user). Tools like Radian 6, Sysomos, Klout help in identifying these potential influencers. This level is hard, time-consuming and relatively expensive but has a higher pay-off .

Forrester’s User life-cycle

Forrester’s User life-cycle (Ref. Figure 2) categorizes users into 6 groups: Creators, Critics, Collectors, Joiners, Spectators and In-actives. This user life-cycle maps to Level 2 mentioned above as it can easily group users primarily based on their age and characteristic alone. But this does not tell you how many creators or critics in a particular age group posted only once. It is also not possible to determine how many of the joiners became inactive after the first login.

User Profiler

The profiler designed here (Refer Figure 3) categorizes users into 6 major groups: Joiner, Information seeker, Active/Dynamic, Responder, Creator and In-active. It is possible to map these groups to the Forrester’s user life-cycle groups with the exception of the Collectors group.

What do these categories mean?

Joiner: Moment a user registers they enter the Joiner level and take a default maturity of 0

Info Seeker: Users who seek answers to questions are the information seekers. Quality of the question decides the users’ maturity. Typically this ranges between 1 and 3

Active / Dynamic: These users actively seek information and respond to others in the group regularly. The maturity for these users range between 2 and 3

Responder: Responders are mostly users within the organization or industry experts. They either have the information or the resources required to secure the information needed for information seekers. Maturity lies between 2 and 4

Listener: These users log in and search for answers but have not posted anything yet. Mostly spectators. Beyond a certain point the lack of user activity except for login’s is either a BOT or Information collector (RSS)

Transition: This is the phase where the users tend to break the routine and shift to other sites or become inactive. One possibility is that the user has found what he/she was looking for

In-active: If the user does not have a single login in a say 1 year, the user is in-active and their maturity score drops to 0. This bucket helps identify and purge records easily for maintenance & performance purposes

Creator: Top management, visionaries and industry experts who are capable of generating interest in the public are Creators. Default maturity is assumed as 5.

Please feel free to get in touch with me if you have any queries or have suggestions around improving the above user profiler.

The aim of this blog was to develop a framework to assess the impact/influence of social media on the user(patients) for those healthcare providers who have invested in it. I am going to split this blog into pieces and the first part is going to build the foundation for the same.

The Common Definition for Social Media:
Social media is the creation and sharing of user-generated content. The content is nothing but unstructured data like dialogues (text), images, audio & video, etc. as a result of interaction and participation between individuals and/or groups.

There are eight pillars one needs to understand clearly to build a strong foundation. They are:

1. The Definition – How does Healthcare define Social Media?
2. The source – Social media sources for Healthcare
3. The Influencers – Who or What can impact/influence the outcome?
4. Trends – Where is the Healthcare industry headed with respect to Social media?
5. The Competition – What is the competition/peers in the industry doing to capture the users attention in the social media space?
6. Monitoring – How is Social Media monitored by the Healthcare companies?
7. What can hospitals do – with Social media
8. How to do – How can hospitals accomplish / get into social media?

Before expanding on the above eight pillars, a few facts on why this topic is relevant today and needs attention.

81% of all Cyberchondriacs have looked for health information online
17% have gone online to look for health information 10 or more times
86% satisfied with their ability to find the information they want online
85% believe the information they found is reliable

(SOURCE: The Harris Poll, Harris Interactive)

Moreover, social media has increased the awareness among users by letting them look up the internet for a variety of things like, facilities provided by a hospital, review about doctors, experience of other patients like themselves, comparison between treatment options given by doctors for the same disease, etc. This has put direct pressure on healthcare providers to not only excel in their services but also involve themselves in the social space.

Starting with the definition, expanding the eight pillars from a Healthcare providers perspective:

The Definition:
This is how Healthcare defines social media, “People trust ‘a person like me’ more than authority figures from business, government and media”, “Seeking ongoing dialogue, not one-way advertisement” and “Trust, transparency, openness, honesty”.

The Sources:
The social media sources for Healthcare not in any particular order.

• Blogs – DiabetesMine, HealthMatters, WebMD, NYT HealthBlog
• Social Networking – OrganizedWisdom, PatientsLikeMe, DailyStrength, JustAskANurse
• Microblogs – Livestrong, Stupid Cancer
• Podcasts – Johns Hopkins Medical Podcasts, Mayo Clinic
• Forums – Revolution Health Groups
• Social Sharing – Flickr, YouTube, ICYou
• Online events/campaigns

The sources mentioned above do not include sources like newspaper/journal advertisements, patient education, etc. Only the online media is in scope for this blog.

The Influencers:
The key influencers in this industry: Doctors, Patients, Patients relatives & friends, Hospital staff, Insurance agencies, Awards and Care givers.

Though these look pretty obvious, it is the word-of-mouth comments & reviews left by these people on the internet that has impacted the hospital most. A simple example, an eye care hospital (name cannot be disclosed) in India started off very well till a time came when a series of mistakes made by their doctors due to negligence changed the tables. Today, the patients do not even want to check if the doctors have changed or not. It will take another round of influencers to change this outlook about the hospital.

Trends:
When we say trends, its more around understanding what the leaders and emerging healthcare providers are doing in the social media space to influence and gain the visibility among the internet savvy public. Some of the interesting trends observed are below:

• Information sharing – Doctors come together to share research, exchange observations, and support one another
• Engaging e-patients – Connecting patients with similar disease processes, ailments, or conditions
• Rating – Of doctors, hospitals, service providers, etc. online
• Targeted search – Connect with patients and potential patients
• Managing a conversation – Q&A with surgeons
• Convergence with personal health records – Reminders to buy prescription medicines, Online personal health diary
• Location-based social media “check ins” – Inova Health System offers flu shot deals

The Competition:
The first mover advantage… and a big hit! It’s all about thinking differently and that idea clicking with the public. Few instances where such innovative/unique thinking resulted in positive visibility for the healthcare provider:

• Broadcast Emergency Room (ER) wait times every few hours across their Twitter feeds
• Seattle’s Swedish Medical Center, along with Mayo Clinic drew attention for hosting the first-ever “Sleep up.” The event, consisted of an all-night live stream following a patient undergoing sleep disorder testing and a Twitter Q&A session with physician sleep specialists, drew 10,000 visitors
• Live-tweet surgeries – Personally this is my favorite
• Interactive fitness program called “Fitfor50″ that incorporates video, Facebook, Twitter and user stories
• Disaster alerting and response
• Social media chats
• Daily health tips

Some of this have become common and no longer a differentiating factor. Example, health tips and social media chats. Users expect this by default though social media penetration by healthcare providers is still in the nascent stage.

Monitoring:
Currently there is a mix of in-house and outsourced social media monitoring and analysis. The major downside to this is setting up the platform for the users to interact and abandoning it. Here abandon means the lack of constant monitoring. When a patient posts a question in a Q&A platform setup by the provider, he/she expects a reply. When that does not happen, negative influence creeps in and the healthcare providers reputation goes for a toss.

Buzz vs. Prevalence analysis on datasets is one of the commonly found monitoring mechanism in the healthcare industry.

What can hospitals do with Social Media:
Few of the benefits healthcare providers can reap by getting involved in Social media are below:

• Help progress medical research when physicians start to interact more often with more social information available in hand
• Publish health news first (alerts, awareness, crisis communication, health hazard information, etc.)
• Monitor hospital reputation
• Introduce new products and service lines
• Strengthen patient provider relationships
• Bring patients with similar problems together
• Physician opinion sharing
• Enable better marketing and communications efforts

How can healthcare providers do this?
In simpler terms, where can they start? Based on my analysis, some of the key pointers are below:

• Dedicated team to monitor social activity
• Integrate marketing activities with social network
• Tools to monitor social activity
• Encourage physician to interact in social forums
• Dedicated team to check and respond to users grievances online
• Setup dashboards for executive team with metrics for social media monitoring

Here goes a listing of 230+ social media tools for extraction, monitoring and analysis. This was compiled in 2011 so some of the tools might not available today. You can leave a comment if you find any such tools in the list and I will have them updated in the next version (2012). Before you start looking at the list of tools, its better you get a clear understanding of the various terminologies so you know which tool is meant for what.

Social Media Engagement (SME) tools are communication platforms where users take action and can respond, engage, interact or communicate directly on social networks, including Facebook, Twitter, blogs, forums, etc…Users provide the login credentials and offer OAuth permissions and can Tweet, comment on a blog post or respond to a question in a forum without leaving the SME platform. These are real-time, highly customized dashboards and often offer multiple accounts, a shared workspace for many users and the ability to respond in multiple places with one click.

Social Media Monitoring (SMM) tools, often called listening platforms, are where most social media strategies begin – monitoring and tracking mentions of your brand, products, competitors and industry issues. SMM tools offer countless ways to analyze, measure, display and report findings

Social Customer Relationship Management (SCRM) tools aggregate dozens of types of information from multiple social networks to give users the most complete view possible of each customer.

Social Media Specialized (SMS) tools fall into a handful of buckets. These tools are not stand alone tools or comprehensive platforms, but specialized tools focused on on analyzing and optimizing one aspect of your social media efforts. The intent is to use these tools to supplement, not replace, other tools. Some of these tools are available on SMM or SME platforms, for example, you can view Klout scores within Seesmic Desktop. These tools can be broken down into the following categories:

• Influence – Klout, PeerIndex, TwinFluence, Twitority, Zuberance
• Conversion tracking – Argyle Social
• Twitter analytics – Bit.ly, Tweetreach, Twitalyzer
• SEO – Hubspot, SEOMoz, Raven Tools
• Xefer
• Ad.ly
• Soovox

Social Media Content Management (SMCM) tools facilitate the creation, distribution, optimization and management of social media content. Each tool facilitates at least one specific activity. For example, Timely posts your tweets at a future time to achieve the most reach.

Click here to download the Excel. The price mentioned is only a ball park as it can vary from time to time and client to client. Any ratings mentioned by the author are based on very small in-house research and might not reflect true ratings.

Tools like Klout, PeerIndex, Kred[1], ProSkore and EmpireAvenue have been making noise recently and Kred’s the new social influence tracker on the block. Klout has been around for a while and has gained enough visibility & traction and hence the focus on Klout. All these tools use a series of algorithms to gauge the users influence in the social world. Off late, this is used by HRs during recruitment process and Sales & Marketing professionals during campaigns.

Influence score is similar to having a credit score. Recently at a “Fashion’s Night Out party” at Bal Harbour Shops in Florida guests had to have a Klout score of 40 or higher to gain entry. Samples are given out to users with a good influence score. This article is about understanding what the score comprises of to improve your social score.

Parameters that influence the Klout score to a significant level are captured based on the analysis of Klout for about four months between May and September 2011. One objective of the study was to maintain a consistent Klout score of 30+ for two months. The result of ~250 tweets and study of ~100 twitter handles, gives you some of the parameters that impacted the Klout score. Twitter handle primarily used for this study was _jk82.

Klout:

Influence matrix with notes

Klout measures the user’s influence across social networks based on the network size, content created, and how others respond to your content. Klout scores ranges from 1 to 100 with higher scores representing a wider and stronger sphere of influence. Klout’s 4×4 influence matrix plots the user’s participation style based on the scores obtained for True Reach, Amplification Probability, and Network Score. To even out errors in calculation, Klout factors the users influence in multiple social media accounts like Foursquare, LinkedIn, YouTube, Flickr, etc.

 

Parameters that have an impact on Klout score

Re-Tweets (RT):

Re-tweets by influential users (good Klout score). Making these influential users to re-tweet you is not easy. So identify this bunch of users who do nothing but re-tweet in exchange for being re-tweeted. Mostly this group tweets quotes so it is easy for you to re-tweet and is less suspicious. Caveat: Do not over do it.

Mentions (@):

Mentions by influential users increase Klout score. Tweet questions that will make your friends reply. Skim through random twitter handles with a decent Klout score (use Trending Now as a starting point) and search for controversial tweets. ‘Reply’ with your disagreement or another controversial tweet, and rest assured you would have your mention.

Unlike Re-tweeting, this one requires some hard work. Influential users know what you are trying to do and do not respond so you will have to do a lot of replies to hit the jackpot. However, Klout does seem to factor in all this effort so it is not a wasted effort. But your participation style might change.

Quality of users Followed

Matrix score

Following active and non-spamming user handle is critical. Tools like Twitoria can help identify inactive users. Install the Klout extension for your web browser and see the influence scores of all the users you follow. Remove users with a very low score, say less than 10. Klout takes into account the influence score of your followers so try and follow more Twitter handles with a score of 80+.

Having a high follower base does not necessarily mean a high score. One can hit a score of 30+ with an average of 50 people following as Klout believes in Quality too and not just quantity.

Following: Followers Ratio

This is not critical but at one point during the study, decreasing the number of users followed to a number below the number of followers gave better spikes to the Klout score. As there are too many parameters involved at any point in time, it is difficult to gauge the effect of this one factor keeping all others constant.

Hash Tags (#):

Hash tags are very important when it comes to “amplification”. Keep a tab on trending topics and re-tweet or comment on tweets to amplify your reach. This is one of the simplest things to do so if required, Google that topic and pick out the best news item or a couple of images and post it with the same hash tag. This will enable other users reading the trend to click on your tweet and even re-tweet.

Links in tweets:

Have a generous list of tweets with links to images, posts, news items, etc. When others click on the links, Klout tracks it. It is good to have a Bit.ly account setup and have the API key supplied to Twitter so you can track the clicks too. Tracked a few twitter handles with fewer than 25 tweets but with links to some video or pic in each tweet and their Klout score was hovering in the range of 15 to 25.

Regular tweets:

Logging in regularly to tweet seems to have an impact on Klout score though minimal. Use tools like Timely.is, Bufferapp.com or Queued.at to schedule tweets. Primary objective is to keep inactivity at bay. These tools can be used to update the status in multiple social networks so if you have classified your user base between social networks like Friends in Facebook, Newsfeeds in Twitter, etc. then this can do wonders to your score. Never hesitate to send thank-you tweets as this increases the engagement score.

Linking multiple accounts:

Linking multiple social media networks is good as it can balance out any error in the calculation of influence. Not tweeting but staying active in another network say Facebook can keep your inactivity at bay as Klout will now search all your linked Social networks to provide the influence score.

Do not go overboard and link all accounts unless you have a good activity going on in each of them. For example, if you are not going to post updates in LinkedIn or YouTube, or there are not too many users looking at your profile, your score can go down. So be wise will linking multiple social media accounts.

Lists:

Create lists with users who have a good influence score or those who tweet interesting stuff and name the list properly. This will make other users follow your list. If you have interesting posts, then others might include you in their lists and more people might start to follow you through those lists. More the lists you are present in, higher the score.

 

Let me know your experiences with Klout if you decided to Game it! I definitely would love to hear how you fared and what more parameters can create that extra bit of difference. Happy fun! and Happy Gaming!

---

[1] Kred bases influence on community & reach, including 4 years summary of tweets.