Sentiment isn’t just for Sympathy Cards!

Posted: May 12, 2013 by Ian Gee in Big Data, Human Resources, OD, People, Social Media
Tags: , , , ,

sympathy card 1

By Ian Gee

We now live in a world with so much data readily available that I sometimes think it is hard to know where to start! Little Data can be intimidating let alone what do we now do with Big Data! My last blog post ‘Making the Soft Stuff Hard’ was an exploration of the challenges this places on our practice. Your comments have really stimulated my thinking. I want to take this exploration a bit further and look at how the web and social media might offer us some new opportunities and possibilities as well as raise some challenges, in terms of data gathering and its use.

In particular, I want to explore the use of sentiment analysis in OD and how it can potentially help us to understand what is going on inside and outside the organisation. Sentiment analysis can be best described as ‘opinion mining’. It’s a software driven process that analyses text and identifies feelings, reflections, likes and dislikes. At its most sophisticated it can give you a very good temperature reading and a good sense of what peoples attitudes are to particular issues. It provides you with an understanding of the judgements people are making, their inclinations, passions and opinions.

Sentiment analysis in OD is nothing new. As practitioners we have always done it through reviewing the data we gather from interviews, observations, focus groups and surveys. Trying to make sense of the data, above and beyond the literal content we search for opinions, similarities and differences. We then make, what we hope, are skilful understandings and interpretations. These ‘data bubbles’ as I like to think of them, have been of immense benefit to me in my career and I hope to my clients as well. Now though, with the proliferation of social media and the web, data is everywhere and not just what we choose to focus on and systematically gather.

Most company intranets have the facility for employees to comment and contribute. In progressive companies, discussion and even dissent are actively encouraged. This generates masses of data and information. But what do we do with it other than watch it pass by like a news ticker? As an OD practitioner, challenged with developing large company interventions, my question is how can we make best use of this rich source of what is on the surface can seem to be random data or chitchat? I don’t believe we can simply ignore it and carry on as we have; we do this at our peril. At the same time, the sheer volume of data can be daunting. It is usually made up of long threads of comments covering multiple issues and usually across numerous platforms. To start to analyse it manually can seem to be bit like trying to read the Internet! This is where I think sentiment analysis provides an answer.

Here is a personal example of where I used a sentiment analysis tool to good effect. I was recently working on a global project where I was not at all sure whether or not some of the key players were truly supporting the changes or not. I am sure you recognise this as a classic issue of stakeholder alignment. Most of us have faced this at some time in a transformation project. As an experiment, I used a very simple and free web tool to analyse a large number of emails I had received from the team and was very interested in the results. By looking at the words and strings of sentences the tool highlighted the fact that the majority of the statements were indicative of people sitting on the fence and waiting. My intuition told me we had alignment issues, the sentiment analysis gave me the data I needed to have conversations with the team and find out what was needed to get them fully on board. Now you could argue I might just as well have trusted my intuition and ‘held up the mirror’ based on what I was feeling. However in a ‘High Tech’ environment, being able to share the results of the sentiment analysis made for a much more fruitful, useful and interesting discussion.

Apparently there are even more sophisticated tools that will analyse anything from tweets, blog posts and discussions both on the company intranet and also the web generally. I think for my own practice, when I am next asked to get involved with a company wide change programme, I am going to recommend that we use both internal and external sentiment analysis as part of the initial diagnosis and continue to use it, at given periods, to see how sentiment shifts (hopefully I a positive way!) as the transformation progresses.

On a final note I am assuming you are all familiar with and other similar ‘trip advisor’ type-sites for business? Have you ever used these sites to help you build a case for change?

I would be very interested to get your thoughts on the use of tools like sentiment analysis as a support to change and transformation. I am wondering if this is this one of the ways in which our practice of OD can be advanced and made more relevant and interesting? If you have any websites demonstrating tools to share that would be great. See below, for a couple of them from my friends Tim and Matthew. Over to you all now!

  1. Natalia Cuervo says:

    Ian. This is a great article ! This is the first time I heard about such a powerful criteria, I’m currently building an EVP and for sure it is a great way to gather quality data!!! So far already run some searches!!!

    • Ian Gee says:

      Good luck with your EVP Natalia and do let us all know what difference doing some sentiment analysis makes. Be good to know how your client responds and what they make of it. Keep us posted 🙂

  2. Lesley Clarke says:

    Fascinating Ian. You’re right. As OD professionals, our judgement interprets what we’re picking up on feelings about issues and that becomes finely tuned if a change programme is long term and highly sensitive and complex (e.g. my last project affecting the terms and conditions of 5000 staff – positive and negative change). My proposals and the emphasis on elements of the package was influenced by my interpretation of feelings. To have software that can assist in judgement on this would be very helpful!

  3. Dee Ortner says:

    Ian, your blog post is (once again!) provocative. Seems as though the technology looks for patterns, trends, and consistency. All good, and generally very meaningful. Yet, sometimes the real sparks are in the outliers, and only become part of the reality when others join in.

    Real opportunities or significant problems are buried in the noise. Without (and even with) human intervention, these can be overlooked. Another firm capitalizes on the opportunity and/or space shuttles burn up. Who’s listening – and how are decisions made? Dee

    • What a powerful way of using data to back up intuition. I’m also interested in what happens when you get surprise responses that don’t match your intuition – also very helpful in this profession. Thanks for sharing this. I’d like to suggest it to those of my clients who only believe the numbers!

    • Ian Gee says:

      Thanks for this Dee and you have made me think about the long tail of data and wonder how we can make sure this is not ignored. I am sure as technological sentiment analysis develops we will no doubt have tools that also highlight the weak signals, surface these and help us explore them. If anyone has seen any tools that help with this be great to have them shared on the blog.

  4. Reb Veale says:

    How encouraging that you are demonstrating how we can explore evidencing the substantive value of feelings in OD. Great business leaders ‘get’ that motivation is an emotive, not a cognitive function.

  5. […] post is written in response to an external blog post (Matthew Hammell’s very thought-provoking “Sentiment isn’t just for Sympathy Cards!”), is somewhat cheeky, but I hope that I can go on to demonstrate that an IT-based response to the […]

  6. Thanks for sharing Ian. We have been using sentiment analysis in our jams to show positive negative or neutral sentiment in the conversation threads. I was at a lecture at Imperial college last week, and a particular example made me think on sentiment. They had been analysing sentiment during Tesco’s horse scandal and sentiment got it all wrong – the explanation was- machines don’t get sarcasm.

    • Ian Gee says:

      Thanks for your comments Julia. I agree about machines and sarcasm! But I also bet there is someone somewhere right now working on an app that will sort this out! 🙂

  7. Ian Gee says:

    Our friend Tim Gorree has kindly reposted this blog onto his website.It has already attracted over 1200 views and been retweeted many many times! He is one of the worlds experts in the fields of gamification and virtual worlds. As well as the comments on this blog I think you may all find his blog of interest so here is a link to it. Read and enjoy! Ian

  8. Ahmed Tahir says:

    “Sentiment Analysis” – I am glad to increase my educated vocabulary with this term. Thanks Ian. It was a very interesting and well-articulated work.

  9. Lisa T. says:

    I like this concept! We always think of surveys as the way to capture objective data, but scaled rating approaches generally limit the ability for people to express how they are feeling. With sentiment analysis, we can still capture a large degree objectivity while people aren’t editing themselves quite as much.

  10. Bimal Rath says:

    Good one Ian–and very useful for leaders in a world which is increasingly driven by sentiment post results (read anti capitalist movements, greening initiatives and the middle east coups). At some stage it will be interesting to experience (and I’m sure it will happen)–along with comments around sentiments from people their actual heartbeats and body temperatures at the time of experiencing a phenomenon and reacting to it.. real sentiment ..and the body is the best transmitter, rather than the mental process.

  11. It’s hard to find experienced people in this particular subject, but you seem like you know what you’re talking
    about! Thanks

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s