How can social listening be made less noisy?

Social Listening - Use of Rules Vs Machine Learning

Signal and Noise

Hewlett Packard has a filtering problem on their Twitter channel. The tech giant’s customers use the terms @HP, #HP, HP to reach out to them via tweets. However when Harry Potter fans submit posts on Twitter they often also use the terms @HP, #HP, HP to refer to their favorite Harry Potter movies and books. Twitter users describing the engine capacity of their cars often use HP as an abbreviation for horsepower. Hewlett Packard’s default listening around the terms @HP, #HP, HP makes for a cluttered inbound feed that contains multiple thousands of posts per month that are of no relevance. Chase Bank has to contend with posts from doting teens tweeting about boyfriends named ‘Chase’. They also hear tweets about people that are seeking to ‘chase’ their dreams. The team covering the Twitter @United handle for United Airlines always know when the English soccer team #Manchester #United have a game, and when there is big news regarding the #United #Nations.

Companies have a choice of technologies that they may use to separate the signal from the noise.

Rules

The use of rules - this approach may include the use of Boolean operators (such as AND, OR, NOT) - is a popular method that most social media monitoring systems employ to enable users filter inbound data. Scripts are developed to evaluate and determine whether posts or comments containing certain keyword patterns should or should not be included in a data stream. Going back to the HP example, a basic script might be something like: Include (HP or #HP or @HP) NOT (Hogwarts). Hogwarts is a popular term used in Harry Potter books.

(Steps necessary to create rules for Radian 6 (Saleforce’s social listening tool.))

Machine Learning

The machine learning approach is more simple. It does not require the user of the system to craft complicated scripts but rather relies on the system to understand attributes and patterns around text. For example, take the need to refine Hewlett Packard’s Twitter stream in order to identify posts that are actionable. A machine learning algorithm would analyze the posts that have been replied to (i.e. acted upon) to draw up a pattern of the posts that it characterizes as actionable. For example, it would see the pattern that posts containing #HP plus hogwarts tend not to be replied to by the Hewlett Packard support team, so the system would learn that  these posts would be identified as not actionable. A similarity index is be used to compare future posts against the pattern that has been built up and a determination is made as to whether any future posts should be classified as actionable or not.

Email Similarities

We all like to stay organized. Suppose you’re working on project called XYZ, you might want to store all the emails regarding project XYZ in a particular folder. That’s a drag to do manually. So you’d want to create a rule in your email application to automatically find and move those messages relating to project XYZ to the appropriate folder. In this type of example the approach for developing a rule in your email system (similar to using Boolean operators) works just fine.

But maybe the plot thickens. Let’s say that it emerges that two of your work colleagues have updated the signatures that they place at the foot of the emails they compose. As part of their departmental responsibilities they indicate that they’re “...working on Project XYZ” in their signatures. Darn. Since the rule you created listens for mentions for project XYZ, now you’re getting all of the incoming email from these two colleagues, not just those that are specific to Project XYZ. No need to worry too much though, you can go back to the rule you created and add a condition to not include emails where the text “working on XYZ” is present.

When the data that you’re working with is simple, known, and easily defined, a rules based approach may work out OK. But in order to keep precision high, each rule needs to be edited each time you encounter an exception such as the one relating to “working on XYZ” in the email signature. The issue with a rules based approach is that can become complicated and scripts may become difficult to maintain when listening for a broad range of ambiguous terms.

Controlling Email Spam

Google entered the market for free web based email in 2004. The space was already packed, with offerings from Yahoo, Microsoft (Hotmail), AOL, and many others. At this time, spam was a bigger problem than it is today for many email users. In order to be successful, Gmail needed to lure users away from entrenched competitors. One of the differentiators that separated Gmail from the crowd was its improved ability to handle spam. Google vastly enhanced email spam filtering by using advanced machine learning methodologies. When a Gmail user clicks the ‘Report Spam’ button, the Gmail spam algorithm logs the message as spam. This signal is centrally recorded so that others in the Community of Gmail users will be less likely to receive that spam message. Importantly, other emails containing similar attributes and characteristics would also be marked as spam. (More on Goggle’s use of machine learning in controlling spam)

The Future for Social Listening

Going forward it’s pretty certain that the volume of traffic (and as a consequence the noise) on social networks will continue to increase. In an effort to filter engagement channels, some companies have created custom handles such as @HPSupport. Creating a rules based approach around these custom handles may be an adequate solution to zone in on some of the outreach from customers. If companies are really intent on listening to and participating in conversations containing obligations and opportunities relating to their brand outside of the narrow lens of a customized engagement handle, a rules based approach is unlikely to be feasible. A machine learning approach similar in use to the one used to solve email spam seems like it will become a necessity. Machine learning doesn’t carry the burden of developing and maintaining complex scripts. The machine learning experience is also more simple and potentially more precise. Rule-based methods don't track changes and don't improve based on patterns of usage. The built-in flexibility around machine learning may be facilitated by a graceful workflow where team members engaging or rejecting (effectively marking as spam) those posts that are not relevant teaches the system what should be listened to, and what is noise.

---

Author: Steve O’Donoghue, Solariat Inc.

SocialOptimizr (a Solariat product) focuses on machine learning and natural language processing to bring leading edge social listening, filtering, and prioritizing to customers.

Read more ...

Notes from Social Customer Service Summit Nov. 22nd, 23rd 2013

The Solariat team just returned from a very successful Social Media for Customer Service Summit in New York last week:  http://events.usefulsocialmedia.com/customerservice/  Big thanks to the organizers, UsefulSocialMedia, for putting on a great event!

It was fascinating to see how much the market has grown in the past year - brands are clearly moving and fast.  There were lots of stories and stats all pointing to the fact that brands and consumers are really discovering each other in social, volumes and intensity are growing.

Our view is that social for brands is fast becoming a big data problem.  While we were confident we were right (anything data rapidly becomes big data in this modern world), we were pleasantly surprised by the fact that this is now widely recognized.  And as a result, this is a market in need of the next generation of technology.  Real technology, not just simple listening and publishing tools left over from the early days of manual marketing efforts.  Scale is the fire breathing dragon just around the corner.

Brands spoke about sheer volume (one astonshing stat about 5 tweets per second...), about the need to triage and sort, about the need to respond accurately and quickly...expectations are rising.  There's just no way that brands can scale with social CSR headcount.  While technology isn't the answer to everything, it's CLEAR that brands are going to have to take a serious look at technology in order to have some handle on the growth and complexity of this medium.

The other big theme we resonated with is that social has to be thought of as a channel just like voice, chat and email but with some important differences.  For starters it's public.  But more importantly, consumers feel a sense of entitlement in social that they don't with other channels.  It's where THEY are  and the brand has to play by rules they don't get to control.  One implication of this is that the demands coming in are broader than what usually gets encompassed by support.  Social is a strong alloy of marketing, care, support, service...

There were plenty of other interesting takeaways about the social channel - all of them pointing at the seriousness of the data challenge.

1.  Social tends to be bursty and fast.  Events hit social FAST and the volumes go up fast.

2.  It's an escalation channel.  For now, a lot of consumers are heading to social after they have trouble in some other channel.  This doesn't seem to be the case for some industries, like airlines.

There was much, much more.  We at Solariat would be delighted to talk to you about your particular experience.  There are exciting and challenging times ahead!

Read more ...

Technology and Social Media Customer Service

By now we’ve all seen the statistics - social media customer service has arrived.  It’s clearly happening.  The only issue now for brands is how to make this work effectively and efficiently.  Social customer service is not the same as community management.  It’s really about accepting that social is one more channel and that it has to accept the same operational requirements and measurements of other service channels.

Social is of course, a very different kind of channel.  It’s public and it’s noisy.  That means that brands have very different kinds of challenges even if they have to achieve the same kind of performance standards.

Every channels goes through certain evolutionary phases. First, there is just the basic mechanism, like setting up the phone lines and email addresses.  Then there are the scaling problems as the volumes go up like how do you handle calls in the thousands versus the hundreds.  As the scaling problems get addressed, the new competitive frontier turns towards optimization.  The brand can handle volume but can it handle volume well.

We all know that the rate of technology evolution is faster and keeps getting faster.  Voice-based service systems evolved over many years, email emerged and grew faster...Social is growing much faster and so brands are going to be pressured to innovate.  Stop-gap and shallow technology solutions are going to become obsolete in very short time horizons.

At Solariat, we decided to take on the big challenges right away.  To accept that customer service in social is not just a BIG DATA problem but a bad and noisy big data problem.  The BIG issue with BIG DATA is making sense of the data.  It’s even harder with social because it’s so diverse, noisy and dynamic.  That’s why the old social listening tools are just not up to the task.  You can’t just listen with keywords.  

And even worse social demands speed in the response.  Brands have to sense, adapt and respond at the speed of social.  As consumers discover this channel, the range and volume of the inbound will grow and brands will have to be ready to respond intelligently.

Our conviction is that real technology is needed NOW. Technology that listens smartly.  Technology that actively assists the CSR in building responses.  And reporting that matches what service managers expect with other channels.

And all of this delivered in an integrated platform that fits hand to glove the workflow of real CSR’s.  That’s the challenge we have taken on at Solariat and SocialOptimizr is our solution.

Read more ...

The future of Customer Service: More Mobile and More Social

Who get’s excited with a new desktop computer these days? Eh, nobody. How about a laptop? OK, maybe a few. Shiny new phones and tablets though get the juices going. Yup, the world has gone mobile. People are spending more and more time interfacing with devices they can hold in their hand or comfortably carry anywhere. OK fine, so how will that that change customer service? Well if your customers are the everyday consumer type (B2B is a different story) probably quite a lot. The reason being, this real-time mobile world brings with it a big slice of time spent on social media, up to 30% according to this Nielsen article.

Will this mean that rapid growth in social customer service will continue to grow? Oh yes it will, for a bunch of reasons:

It’s here, it’s now, it’s handy

Twitter, Facebook and the other multitude of social networks are the apps that consumers are likely to have persistently open. This is where they’re “hanging out”, so this is where they’ll be more likely to launch their service requests. Like it or not, our society has become distracted by our digital mobility and real time access to information. Things have become prioritized based on the distance they are from our noses. If we’re standing at a bus stop and thinking about how irate we are with our cable company, we don’t have to make a note to call them later. A quick tweet or Facebook post to let them know we’re not OK, done. At the airport and upset that as a premier customer you got bumped off a flight, or annoyed while waiting too long at your local bank, same deal. The new norm will be to vent and demand better service through social. Then on to the next distraction.

Can’t even find the darn phone number!

Searching on mobile is not so easy. Finding that phone number for your cable company, airline, or bank via your handheld, probably not going to happen. And anyway if it’s not critical do I really want to wait and navigate through a convoluted decision tree. Hash tags are easy though. Why not just use one of those? Less phone calls more social.

Chat? No so much

Ever tried synchronous online chat on your mobile device? Not so easy. Active chat conversations work best if you’ve got a full-on punchable QWERTY keyboard. And anyway you’re probably doing fifteen different things while mobile. How’s about a social post with details of the issue? Hey company, fix the problem, or post me back when you know what you want me to do next.

Make it 24/7, and make it snappy

Everything else happens quickly in this mobile world. Shouldn’t a response to my issue be the same? Studies show that a majority of social users expect a reply to their social post within an hour (that’s 60 minutes folks). So say good bye to those weekends off if you don’t want a backlog of missed posts when you come in Monday morning.

And what about forums, and self-service knowledge bases?

Forums the self-service solutions will always have legs. As more and more of these products are elegantly designed for mobile consumption they’re popularity will likely persist. But, like the company telephone number, searching, and navigating for the precise content needed is still often very challenging.

But wait, does that mean the other channels will go away?

Course not. Not at all in fact. Business-to-business customer service will see the fewest changes. Most business customer issues are submitted and resolved by folks sitting in their cubes while on the phone or online via “traditional” non mobile devices. Social for B2C will for sure grow at a rapid clip though as more companies realize that it’s in their interest to attend to it, and as consumers realize how easy it is to use they’ll use it more. And don’t let anyone fool you, problems, big problems, can be solved via 140 character exchanges. The limitations on the length of the post can be overcome but replying with links to knowledge articles or instructional videos and the like. But, obviously, once an issue passes a certain threshold of complication, or necessitates the exchange or private details, noisy social customers can be scooted back into those oh so calm and serene channels like phone, email and chat. Namaste :-)

Read more ...