Monitoring & Analyzing Social Media

SM2 from Techrigy Gets Emotional: Sentiment, Tone and Emotion in Social Media

SM2 has always had a sentiment analysis tool designed to help users track positive/negative opinion on brands across conversations in social media. This week we took the sentiment analysis to an entirely new level by adding tools that measure tone and emotion.

Tone measures the overall tone of a social media conversation on a scale from very positive to somewhat positive to neutral to somewhat negative to very negative. With this tool you can now view results sorted by these criteria and combine them with other metrics like Popularity. If you are doing reputation management, for example, you might want to focus on high Popularity sources that are very negative for your initial engagement efforts.

Emotional Tone is a different kind of look. We offer the ability to view results that show strong emotional and even physical responses:

The emotive states we cover include:

Anger, Sadness, Social, Family, Friend, Anxiety, Bio, Body, Sexual, Ingest, Achieve, Home, Money, Religious, Death and Leisure-related. For example, Ingest-related would include references to eating, drinking, dieting, etc. The chart above is from a search on Obama’s Renew America campaign done with the AD Council so it accurately has emphasis on Social and Achieve-related emotional tone. The initiative is a volunteer service so those talking about it in social media are generally very positive and interested in the social and achievement aspects of the campaign.

As we continue to add additional features we’re always interested in feedback, ideas and examples of how you are using SM2 and what would be useful to you. Just shoot a note to support at techrigy dot com.

ReBlog: 42 social media pundit predictions for 2009

Here’s a nice round up of the various blog pundits’ predictions for Social Media in 2009.

Thanks to Joe Pulizzi of Junta 42 for putting this together.

B-B Lead Generation in Social Media

Leads are the lifeblood of any business and generating them is the ultimate goal of any marketing. I’ve been in companies where we rented leads in bulk (total waste of money) and others where we acquired them by networking at events (good but labor-intensive). You can fish for them with SEO and PPC (better because you’ve defined intent on the part of the prospect) but you’re dependent on the search engines and the quality of your site. Finally, you can generate leads through freemium versions of your service or product. This works well if you’re marketing a professional service and you’re providing enough value for a user to trade you contact info in exchange for their freebie. This freemium model is a piece of our lead generation strategy for SM2. However you still have to get to your site to sign up. That’s where social media fits into lead generation

Adding in Return On Investment (ROI) for Social Media Lead Generation

With a tool like SM2 you could find everyone talking about the keywords associated with your product, extract their addresses and/or URLs and try blasting them with some kind of offer. This may sound good if you’re used to the old media broadcast model- but it is a really terrible idea in social media because there can easily be a backlash from participants who don’t like being spammed (which is what you’d be doing- they never asked for your offer). So how do you generate leads in social media?

Try a variation on this process:

• Use SM2 to monitor conversations that contain your keywords, then refine those keyword to more closely target sources. This helps qualify your leads.
• Look at the Author Tags cloud in SM2 to find ways social media results authors are organizing their conversations. You can find clues to additional keywords.
• Build an offering that focuses on those keywords. We offer a free version of our service, others offer discounts, white papers, webinars, etc.
• Create a landing page that focuses on your keywords and the specific interests of your prospects. This is not your home page, it is a page dedicated to gathering information in exchange for your offering. Keep it simple, don’t ask for too much, etc. We only require an email address but we ask for Name, Company, Title and Where They Heard of Us. About 50% offer this info which is high but we don’t make them do a demo or talk to a salesperson, the free version is fully functional and we have an additional goal of building an active user-community.
• Start delving into the conversations you find with SM2 and become a participant. Don’t pitch, participate: respond to relevant Tweets, comment on blogs, join networks. Be transparent about who you are (I often define my user-name as Martin Edic (Techrigy), making it clear that I’m from a company). You’re not selling here, you’re building a reputation.
• Have the link you provide in your Profile or Comments be a direct link to the landing page you built. Many will click to find out who you represent if they find you’ve added value to the conversation.
• Offer up your offer if it adds value to the conversation with a URL in your response.
• Don’t use canned responses.

I realize this sounds kind of labor intensive because it is. Many companies are hiring Community Managers to do this.

So where does the ROI come in? First, social media is exponential. Your comments and tweets may be found and read by hundreds or even thousands of readers, readers who are qualified leads, otherwise they would not be there. Second, people in social media like to spread the word so your efforts go even further. And your conversations usually don’t go away- they have a shelf life and keep on giving.

Many people in marketing are not used to valuing leads. This is incredibly important to determining ROI. A qualified lead is one for which you a minimum required amount of information. Another qualification is the source of the lead. The rented lead has a suspect source, the lead that followed a process to get to you is very well qualified. One might cost $.15, the other $150.00. You determine how much a lead is worth by working backwards. First you know how profitable an average sale is, then you determine how much of that profit you’ll expend to get that sale. You need to know how many leads turn into a sale. It might be one out of ten, in which case you need to generate ten leads to make one sale. Do the math and you can determine how much a lead is worth. Run that against the required investment to generate that lead and you’re much closer to an ROI calculation.

The essential introduction to why Twitter matters and how to dive in

Shel Israel has the best description of why Twitter is important and how to use it. Twitter is gaining credibility as a communication tool now that it is finally moving beyond the ‘my favorite doughnuts are fried cakes’ model.

For example Alisamleo at Socialized has been attending a Search Engine Strategies (SES) conference in San Jose and Twittering about their overwhelming misunderstandings regarding social media.

Exporting data with user-configurable fields from SM2 Results

We’ve been asked by customers and users for the ability to export data from SM2’s results reports in a format that allows users to use that data with other reporting solutions. We’ve always offered the ability to export but the entire record came out as one data field. Now we’ve added the ability to choose the fields within the SM2 search results when you export.

Here’s how to create your export:

Under Reports click View Results. You can sort the results you’re exporting by selecting the Categories pull down at the top. This helps you sort by thing like Sentiment or any rules you’ve set. Once you have the list of results sorted Click the Export To XLS button on the top left. A dialog box will appear with checkboxes for all available fields (currently 31). Select those you wish to export, click Export and you’re done!

Reputation Management in Social Media: SlideShare

Slideshare: SM2 Analysis Guide

Reputation Management with SM2, Sentiment notes and Reputation ROI

The super-hot political season we’re in is a great time talk about reputation management in social media as professional smear artists work overtime spreading rumors about candidates of all stripes. Obama’s campaign has aggressively recognized this with their Fight The Smears site where they immediately publish any new rumors and refute them decisively in real time. I don’t know if they are using social media monitoring to follow what people are saying but I think it’s likely. This kind of proactive reputation defense requires a combination of technology and human involvement.

For example, though we offer a sentiment indicator in our analysis tools it is just that: an indicator. It identifies words and phrases in context that it thinks could be indications of a negative or positive statement related to a keyword in that search. If it saw ‘Obama sucks‘ in a blog post it would likely flag that as negative. This is where the human beats the computer every time however. Using our drill down feature you can read the ‘negative’ statement in context. Suppose it actually says:

‘ Obama sucks down a frosty at a local fast food joint while talking to a smiling group of fans’

The computer thinks that’s negative, any human knows it’s positive and would correct the sentiment in SM2 accordingly. Yes, it’s labor intensive but not as intensive as rebuilding a reputation damaged by an untruth or misconception.

Sentiment and Accuracy Claims

Semantic search offers up the Holy Grail of search, search that understands natural language queries such as:

‘which dealer in Rochester has a blue Civic in stock?’

The amount of things a search engine would have to understand to return an accurate answer to this question is mind-boggling. It would have to know that Civic and dealer in the same sentence probably means a car is involved, that ‘in stock’ is a sort query and that blue is an attribute. Then it has to know that we’re only interested in Rochester dealers.

This kind of thing is why we have to be very wary of claims of accuracy in sentiment analysis. Unless a service is having actual humans read every result you can only use sentiment as a guide to the general direction of the discussion.

Reputation also varies with demographics and you can see some of this in SM2. If SM2 shows a majority of males from 34-50 in the Midwest think Obama is a Muslim (he is not!), then your management has identified a particular demo in social media that requires your attention and some remedial action.

Reputation management is labor and time intensive. It requires real time discovery because distortions can travel extremely fast in social media, the ultimate rumor mill. Like a recent Doonesbury storyline depicting his weary daughter relentlessly scanning the web 24/7 for Obama smears, it requires a lot of attention.

ROI for Reputation Management?

How do you measure the cost of swing voters in a hotly contested state? Of a false product rumor that derails sales overnight? Of not being prepared when a new market sector latches onto your product for a use you never considered? The ROI is based on risks averted which is tough to quantify.

ROI for Social Media Engagement: B-B Lead Generation

The subject of social media engagement for marketing purposes is a touchy one, especially with B-C marketing where consumers are very likely to be angered by any kind of pitching in social media. However, for B-B marketing it’s a whole different story- I know because it’s Techrigy’s primary marketing channel and may become the primary channel for B-B lead generation period. But can you measure it?

You can if you plan a campaign, not a campaign like an ad campaign, a campaign designed to turn your engagement into measurable results. You start by hiring or assigning a Community Manager to a product or service within your agency or marketing group. This person’s job is to reach out to social media via a service like ours (overt plug!) and participate in conversations. That’s all well and good but how do you justify the overhead involved?

You start by understanding how people respond to you. If you’re adding value to the conversation via blogging, comments, Twitters, etc. make sure there is a link associated with your name at all times. If people like you response they are quite likely to click on your name/link to find out what your story is. Where that link takes them is critical to creating and measuring conversions.

Let’s look at an example from my world. There are a lot of blogs and conversations out there about social media and brands. It’s a huge subject with a lot of unknowns because we’re creating a whole new world here. This means people are hungry for knowledge and shared experience. We’re a B-B (business to business) company so my real marketing goal is the generation of highly qualified leads. That’s the end game for me. My target market is all of the people out there trying to figure out social monitoring and engagement.

So my message when I comment is always related to helping my peers, who happen to be my prospects, understand what we’re trying to do and how. Because I stay relevant to the thread I’m in and I only pitch when there is a direct request for information, people are clicking my name/link. That link takes them directly to the free sign-up page for SM2 where they can fill out a simple form to get a free account. Those that do are considered conversions.

Very quickly, the formula for ROI is # of leads that sales can close X average value of sale. Say you turn 5% of sign-ups to paying customers (these are made-up numbers) and your average sale is $5000. That makes a lead worth $250 (5000/20=250).

So how do you measure conversions? You can use unique URLs for your link but that doesn’t take into account word of mouth (WOM) which is an exponential factor in social media. You can ask people in your form how they found you but only a few will offer that info. You can use site Analytics to track referring pages but that has limited effectiveness in SM.

The fact is that there is no totally accurate way to measure the exact ROI for your social media engagement, however it is far more measurable than virtually any form of traditional brand advertising. How are you measuring engagement?

BTW, I’m experimenting with Zemanta in this post, a WordPress plugin that adds links and images it thinks are relevant. Not sure what I think yet…

Some words about keywords in social media monitoring

SM2, like most of our competition, is driven by a keyword set-up you do when initiating a search. We provide a tabbed interface to help you define sets of keywords that will accurately drive your search results. These include keyword categories like Organization, Competitors, Products, People, etc. By using these categories when setting up your SM2 search you prepare the application so it can help you sort and analyze results. We also offer the ability to utilize negative phrases to eliminate undesired results- plus some advice on not using words that are too generic.

Most of us are familiar with the use of keywords in search optimization for web sites and in contextual ad campaigns like Google Adwords. These keywords are the phrases most commonly entered into search engines by people seeking information. Keywords for social media monitoring are a little different because we’re looking for things people refer to in conversations rather than search terms. ‘Buzzwords’ might be a better description.

I recently did an SM2 search on the subject of Al Gore and We Can Do It, his climate change awareness campaign. The only keywords I used were “Al Gore” and “We can do it” and wecandoit.org, their website. These were specific enough to return a lot of results, enough for me to see trends, identify sentiment (highly positive, BTW), get a good idea of geo-location, etc. Other ‘brands’ can require many more keywords, particularly when there are many competing brands, multiple markets, etc.

If you’re finding that you’re getting too few or too many relevant results you can clear your search and start over with a better set of keywords, perhaps removing generic terms or adding a few more specific phrases. And if you’re tracking someone with a common name like ‘Mike Johnson’ you may want to associate his name with something less common like the name of the company he works for. This creates a phrase that kills off getting a lot of irrelevant results.