Social Media Darwinism » Search

Use Facebook Data to Unlock Consumer Affinity

Paul Dunay — Wed, 02 Nov 2011 18:00:48 +0000

When I started this blog back in 2006 Facebook was just getting going but nobody had a clue on when it came to the volumes of valuable data that they would ultimately produce. Today we are talking about almost 1 billion posts on Facebook per day, the same happens on Twitter every 4 days.

When I first saw the way in which Facebook data could combine with Twitter data, blog, forum, search and web data to make a rich profile of your consumer – I was blown away by the possibilities of knowing your customer better and knowing what really makes them tick.

Check out me and Dr Manuel Aparicio from Saffron Technology discuss the possibilities of harnessing the power of Social Data on yBCtv!

Special thanks to George Barker, the man behind the video magic, over at yourBusinessChannel the people who love “telling business stories and getting them out via a network of networks” for producing this video. See yourBusinessChannel for more videos with key thought leaders on the topic of social media and beyond.

Ideal Customer Persona – made with Social Data Some advertising agencies charge you hundreds of thousands of...
Why Your Marketing Needs Real-Time Data Another Advertising Week in New York is upon us and...
Facebook gave us the last piece of the Earned Media puzzle A small relatively un-noticed posting on the Facebook blog...

Related posts brought to you by Yet Another Related Posts Plugin.

Is It Time To Reconsider Sentiment Scoring?

Paul Dunay — Wed, 10 Aug 2011 18:00:23 +0000

In a recent blog post called The Problem with Automated Sentiment Analysis from Fresh Networks, a social media agency, they evaluated a few sentiment tools and their results are quite similar to what we’ve found in a number of our own experiments:

- About 80% of posts are neither positive nor negative.

- Sentiment tools “accuracy” of 70% to 80% is largely driven by their ability to correctly label neutral posts.

- “In our tests when comparing with a human analyst, the tools were typically about 30% accurate at deciding if a statement was positive or negative”

From the blog comments, it’s clear that the companies in this space are doing their best to obfuscate the truth. To some’s credit, they do state that sentiment alone is not enough information to derive any conclusions.

However it’s NOT better than nothing, it’s actually worse than doing nothing because you are getting INCORRECT information.

With sentiment there is no such thing as accuracy, there is only agreement. The technology can’t become more accurate, it can only agree with people more often. And, “sentiment” does not mean the same thing to all people in all situations. You can’t get more “accurate” at “sentiment” because what you are actually talking about is trying to solve hundreds or thousands of slightly different problems with one tool. Until we can map the human brain into a program or electronic circuits, I just don’t think that is going to happen.

I completely believe that having inaccurate sentiment is worse than having nothing. Here is a good example. In posts about “Blackberry” that have been classified 3 different times by hand, about 32% of posts are positive (with a majority vote). When we take that same data set and have each post classified 10 times, now about 10% of posts are positive (with a majority vote). And, if we only consider the posts we are confident in, only about 3% of posts are positive.

So, which is it: do 30% of people like “Blackberry” or do 3% of people, because that’s a BIG difference. Of course, the answer is probably neither because we aren’t actually measuring how many people like “Blackberry”. Unfortunately, that’s how it can be interpreted. Hence, bad information can be worse than no information.

Marketers need to be ware that a lot of these companies say they do monitoring and provide analytics like sentiment but in reality they are really keyword-focused listening platforms with limited analysis capability. If you really want to go beyond sentiment analysis you need to use semantic analysis. With semantic analysis marketers can better understand the conversations about their brand or product category– here is a white paper that compares Semantic vs Sentiment Analysis and can help you make a more informed decision about when and how to use Sentiment.

Why Semantic Analysis trumps Sentiment Analysis For years, sentiment has been a widely used measure...
Is Sentiment Making Brands Stupid? I loved the article by Nick Carr a few...
Social Media – the Next Frontier! Many BtoB organizations these days are spending time and money...

Related posts brought to you by Yet Another Related Posts Plugin.

Why Semantic Analysis trumps Sentiment Analysis

Paul Dunay — Wed, 04 May 2011 18:00:44 +0000

For years, sentiment has been a widely used measure of how customers view a company’s products and services. But sentiment analysis has inherent flaws. First is what it cannot tell you because it only considers a small amount of the available data. Only about 25 percent of posts actually contain sentiment, either positive or negative, which means three out of four posts are neutral, revealing no sentiment, and are effectively being ignored by the analysis. Thus, decisions are being based on what only a quarter of the posts are saying.

Another problem with sentiment is statistical confidence in the data. Simply stated, all methods of sentiment analysis rely on example data that, whittled down, reveals a low level of confidence about the sentiment being identified, either positive or negative. Data with such low confidence is a poor foundation for sentiment analysis.

Bottom line — sentiment analysis is not inherently bad. In fact, for particular types of questions, it may be the right approach. But if you use it, you need to be sure that it’s the right tool for the job, that valuable data is not being ignored, and that the method of sentiment analysis is built on sound data.

A much more statistically reliable approach is semantic analysis — a way to distill and create structure around mountains of unstructured data, such as blog posts, social network chatter, tweets and more, without preconceived ideas of whether or how they are related.

Semantic analysis allows you to cluster different data elements based on similarity, rather than preset classifications such as positive, negative and neutral. This helps you uncover important information like what exactly people are saying about your product or service; where and how they use it; and enhancements or new offerings they’re interested in. This type of valuable information can drive product development, new revenue streams and strategies for marketing, advertising and media planning.

Click here to read an interesting report that digs deeper and compares Semantic vs Sentiment Analysis.

Monitoring vs Analytics [Infographic] After two years at the helm of Avaya’s social media...
The End of Traditional Marketing Paid, owned and earned media. Those three words, often...
Use Social Media to set Smart Media strategies Let’s make the assumption that you’ve successfully integrated your paid,...

Related posts brought to you by Yet Another Related Posts Plugin.

Micro Targeting your Online Media

Paul Dunay — Wed, 23 Mar 2011 18:00:37 +0000

Let’s say you’re an auto insurance company. You want to target micro-segments where people think about buying auto insurance. We know that people buy auto insurance when they buy a car. But where else can we find people who intend to purchase insurance?

Using social data, we have discovered that people tend to think about auto insurance when their kids are going to prom. They think about it then because they’re worried about whether they have enough insurance. A risk mitigation scenario is taking place that you can target with your media plan.

Another “decision intersection” might be when men have had a mid-life crisis and they’ve just bought a big, fast car. If you can learn where those men congregate in the social media arena and how you can place your media in those areas, you can improve the efficiency of your media buys. To answer that, you need to discover where men go to have those conversations and then place your media there. And that media could be in the form of an advertisement or in the form of paid content.

By using social data analysis, including topic discovery, you’re likely to have a much more efficient media plan.

To read about more clever ways to strategically plan online media, check out this FREE eBook on the 5 new tricks for media planning.

How to Optimize your Paid Search with Social Media We used to learn about who our audience was...
Social Media – the Next Frontier! Many BtoB organizations these days are spending time and money...
Paul Dunay joins Networked Insights as the CMO Almost 5 years ago I started this blog with the...

Related posts brought to you by Yet Another Related Posts Plugin.

How to Optimize your Paid Search with Social Media

Paul Dunay — Wed, 16 Mar 2011 18:00:00 +0000

We used to learn about who our audience was by putting cookies on our website and waiting for traffic to come so we could learn more about our visitors. For example, before people visited our site, perhaps they visited espn.com. So we potentially have a large sports audience as a segment of our brand, right?

The search process has become broken because the data landscape has become more fragmented and complex. If we know consumers attach certain keywords to our brand and that certain topics are becoming important to them, we can probably buy those words as search terms before they become expensive, and we can probably own them not only from a paid standpoint but also from a natural standpoint. But how do I find those keywords (see “Why Topic Discovery Beats Search”)?

When I put a search term into Google, typically eight of the top 10 results are social because consumers are creating enough content for that to happen. If I can discover the trends and themes consumers are creating, then I can put myself in the mix by building content to show up in that top eight to 10 results as well. As a media planner, I need to know the top trends and themes my consumers care about, so I can buy those terms and then optimize based on those terms as well.

Consumers today don’t search with long phrases. They search once and if they don’t get what they want, they tend to add to it and keep adding to it until they find what they want. The great thing about using social media data for topic discovery is you can see all of those searches connected together. For example, you can see the result is first “social,” then it’s “social media,” then “social media analytics,” and then “social media analytics paid.” This is what we call knowing the hierarchy of search, which can only be accomplished with topic discovery.

To read about more clever ways to strategically plan media using social media, check out this FREE eBook on the 5 new tricks for media planning.

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.

Why is Google Scared and What it means for B2B Marketing!

Paul Dunay — Tue, 26 Jan 2010 16:23:21 +0000

At a recent dinner with a buddy of mine from Microsoft (who helped launch Bing) – I learned a lot more about the search market and a very important trend that can effect the future of B2B Marketing.

When you think about the search behavior of the average person online – we are mostly conditioned to search in very routine ways like pulling up a Google search page, using the toolbar in your browser etc. And we tend not to stray from that behavior now. It’s almost an unconscious choice for most of us.

But when it comes to search on a mobile device we mostly use very specific apps, like Yelp for Restaurants or Flixter for Movies – where the search result is much much richer – you not only get results that are more relevant but you also get a chance to make a reservation, get directions, buy a movie ticket right there within the application.

If you look at the share of mobile search that the Google app has it is minuscule compared to their share of the browser based search market. This is why Google is so interested in buying apps like Yelp or launching their own phone like Nexus One since they realize this is the year mobile search really takes off and if they just stick to browser based search their overall share will diminish.

OK so what does this mean to B2B Marketers?

While there is no app yet for finding and reading Thought Leadership on your iPhone or an app for finding advice from an Expert in a particularly difficult technology topic (business idea?) perhaps there will be shortly. Maybe apps like foursquare could have a business to business version in the near future. I think as B2B Marketers we need to be open to and looking for those types of opportunities – because one thing is for sure – the devices we have in our hands are going to only get better and traditional search while it may always be around in some form – new avenues on mobile devices are going to open up for us.

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.

LinkedIn Search Optimization (LSO) Critical for B2B Marketers

Paul Dunay — Thu, 07 Jan 2010 16:18:47 +0000

Search engine optimization (SEO) is the process of improving the volume or quality of traffic to a web site from search engines via natural search results. This gave birth to a whole industry centered on optimizing search results for a given term. For B2C or B2B Marketers this was a huge revolution.

But for B2B Marketing fans I think you need to expand your game around search to start include what I am calling … LinkedIn Search Optimization (LSO).

Late last year LinkedIn launched an advanced search feature on their site. And if you haven’t already spent some time plugging in your favorite B2B marketing keywords either paid or organic – you should. The results you get may surprise you.

I would start by searching for terms where you would expect your major thought leaders would appear – if they don’t show up on the first page of the results I would start perfecting their profiles to clarify what they do and what their strengths are using keyword rich descriptions in their bio and the description of their current role. Their headline should also be changed manually to display their expertise not their last job (which is the default). Also, your thought leaders email address should be at the end of the summary area or in the contact field labeled “public”. And you may want to remind them to update their status as often as possible. People in their network are most likely interested in their new projects and professional achievements – why not keep them abreast of what’s happening so they can refer them to a friend.

Personal branding expert Dan Schawbel suggests that your profiles should be rich in keywords that anyone would be searching for. For a good example check out these searches for the terms Facebook Marketing and Social Media. Ok ok yes I was doing a vanity search just to see how I would rank but I was very impressed by the results I got for Facebook Marketing – someone had their profile so well tricked out with keywords that they ranked higher than people who actually work at Facebook.

Clearly there is a new practice emerging here, a practice I feel B2B marketers can benefit from if they focus on it.

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.

Social Search could it be a Google Killer?

Paul Dunay — Mon, 29 Jun 2009 14:39:00 +0000

This morning I typed into Google the search term: VoIP (which stands for Voice over Internet Protocol) and I got 64 Million responses. 64 Million!!

How archaic is that?

Does Google have any idea how long would it take me to go through 64 Million responses? One day we are going to look back at the days of searching and getting 64 Million responses to anything and equate that to a Commodore 64 or Betamax tape.

I know, I know the point is Google is trying to bring me the best responses to the term in the first few pages (if not the first page). But the best responses according to who? Why don’t they know I work at Avaya (who sells products and services in the VoIP space) and therefore I most likely have a very different mission in typing in VoIP?

Microsoft just launched Bing a new search service which is supposed to rival that of Google. To me it looks like Google but in a new wrapper with a few enhancements. This post is not an indictment of Bing, my point is they didn’t take it far enough.

Someone can (and will) connect search to my social profile (say on Facebook) thereby making a truly intelligent search engine that will know where I work, and who my friends are. Then when I Google a term like VoIP – it will know I work for Avaya, it will know my friends and can present me with a way to filter the web based on my social profile – call it Social Search.

Then I will be able to narrow down a search for VoIP down in seconds based on the company I work for, members of my marketing team, friends I have in my network and get a completely different view of the internet that I can’t get now.

24 months from now Social Search should be common place – we should see social tools like Facebook Connect or OpenID which are already out there, used for search. I feel everyone wants to make the search market more competitive (right now Google has 87% of the US marketing and 90% of the European market) so perhaps Yahoo or AOL can revive their search business by going Social!

For more ideas like this please check out my latest eBook called 7 Ways Facebook will Change your Life!

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.