social media

Big Data & In Memory Computing

BT Tower - London - A Monument To The Broadcast Era

Big Data is a term coined to describe the data deluge we are currently experiencing. It’s no secret that we are living in a technology driven society that generates an ever increasing amount of data as we go about our daily lives. “Always on” is a term that is often heard and more often then not, when we are “on” we are creating data about ourselves, our likes and our dislikes, our network of friends both professional and social, and even our travel habits. At the same time, businesses must also retain more and more information to manage themselves more efficiently across the board. In tough economic times there is an ever increasing recognition that organisations must use every single resource at their disposal to get ahead. This results in information and data that once might have been given little attention is now seen as worth its weight in gold if any perceived value can be derived from it.

Looking at the sources of this data, to start with there is of course the sales, operational and customer data that the business collects. On top of this there’s social media data from the likes of Facebook and Twitter including information around friend groups, likes/dislikes and sentiment analysis. There’s web search data, with transactional information or online customer reviews. There’s also data generated by location-based services and data from sensors, moniters and GPS embedded in a growing array of products from vehicles to appliances. I believe if we as advertisors can offer solutions to our clients of how we can process and utilise the growing amount of data available to help inform creative business solutions we could offer real value. To quote a recent Financial Times article

“The challenges are two-fold: First, to recognize the value of big data in mining customer needs and desires, and second, to devise a data management strategy that integrates big data into the front end of the innovation pipeline.” 

So how do businesses harness all the data that is being created and use it to inform their strategy and decision making? The challenge here is being able to process large amounts of data at speeds that make it useful. It’s very difficult to really make use of data in business decisions on an ongoing basis when it takes weeks to gather and process. Large software powerhouses such as SAP & Oracle have been bringing new tools to market for businesses to help solve this problem. In memory computing software is designed so organisations can analyse vast quantities of data in near real time across many sources. Essentially in-memory computing takes advantage of a better understanding of how data is formed and housed and the ever decreasing price of memory (discussed in my Technology vs Advertising post). Instead of housing data on a hard drive, data is stored in a computers memory. Therefore, when it needs to be analysed it is available in near real time. This increased power and speed also means that the computers can handle more unstructured data, important when data can come from so many different sources. On the back of this there would need to be a process for managing and delivering the data in an efficient manner and most importantly, in a way that is easy to understand and glean insights from.

Whilst I think caution must be taken not to let our ability to measure granular details bog down the creative process, at the end of the day, the more you know about your customers and can integrate those insights into your business strategies the more likely they are to improve revenue, margins and market share. Who wouldn’t want that leg up over the competition?

Alex

Colmar Brunton & Millward Brown’s Digital Predictions For 2012

As we rapidly speed towards the end of the year and scurry to get the last bits of creative out the door before we head off on our Christmas breaks, I took some time today to consider the year ahead after reading Colmar Brunton & Millward Brown’s digital predictions for 2012. Whilst it’s impossible to be truly accurate with these predictions, it’s exciting to think about the trends that will potentially be shaping our projects in the year to come. With technology changing so rapidly, alot can happen in 12 months time! Here’s their list:

#1 Gamification Unlocked: Big Brands become even more playful

This concept has close ties with behavioural economics, applying gaming mechanics to non game situations. The trick is to bake them in and do it in such a way that it doesn’t feel like a game. For example, bonus reward points randomly offered on purchases.

#2 Just Tap It! Wide spread adoption of the mobile wallet

We’ve always got our phones with us, we will start to see its functionality enhanced enabling practicality. It could be used as a payment system, a means of I.D, drivers license or even keys to unlock our car.

#3 Virtual Togetherness: TV & Social Media will fuel an explosion in tools, technologies & platforms for interaction and research

People will be able to engage with shows in new and interesting ways. TV viewing habits are often influenced by our friends so the social graph could be leverage as a program guide. Could have an impact on how we view TV Ratings also.

#4 Online Video: Invades the living room

Google TV & Apple TV have already had a stab at this, but as TVs become internet enabled we will be able to access more and more content from the web with it.

#5 Mobile Marketing: Will become more social & local than ever before

Success in this space will combine relevance, location and timing of the content intertwined with social. Care must be taken about appearing intrusive here as people have very close relationships with their mobile phones and don’t take well to being pushed content they don’t want.

#6 Growth: The only App trend that really matters

Looking beyond Apples app store, we see Android market steadily growing. Apps could be pulled accross multiple devices from mobile to tablet to PC and TVs. In App advertising will grow and morph into richer content and video.

#7 Social CPG e-commerce: Tiptoeing between engagement and marketing leads us back to traditional marketing vehicles

Care must be taken not to turn the social space into just another selling opportunity but it may offer the chance to build awareness, trial, sample and coupon.

#8 The Social Graph will generate meaningful data for brand measurement

The social graph generates an unprecented amount of consumer data that can be used both for consumer insight and a real time barometer of consumer opinion.

#9 Regulators narrow their focus as consumers pay the real price for “free” access

With information sharing essentially being the cost of entry to networks, the financial implications of how information is managed and protected may come in to play.

#10 The arrival of Seamless Sharing

The share button will continue to spread throughout the online space, being able to share a peice of content is just naturally expected. Those who produce the most interesting content will gain the most earned media through this mechanism.

#11 China will see “One Stop Shop” convergence of micro-blogging, social networks & information portals

New forms of micro-blogging which integrate facets of social networks are taking off.

#12 Online Advertising: Real-time decision making takes centre stage

We want data and we want it right away. Successful strategies here will involve being able to adapt efficiently to real time information in combination with insight, analytics and creative execution. As a side note, Google Analytics just launched real time stats for your page.

Colmar Brunton & Millward Brown’s digital predictions for 2012 Report

Here’s to an exciting 2012!

Alex

Considerations of Influence

Lapo Elkann - considered a style icon

I just read a post by Noah Brier in which he quoted the below from the SocialFlow blog regarding how we consider social influence which I thought was quite good and got me thinking about how we measure and define influence.

“As we build out digital social spaces, we must not get derailed by metrics of status affordances that have taken center stage. Just because we have easily accessible data at our fingertips doesn’t mean that we have the capacity to model and place a value tag on human behavior. Followers, friends or likes represent an aspect of our digital status, but are only a partial representation of our general propensity to be influential. Keith Urbahn wasn’t the first to speculate Bin Laden’s death, but he was the one who gained the most trust from the network. And with that, the perfect situation unfolded, where timing, the right social-professional networked audience, along with a critically relevant piece of information led to an explosion of public affirmation of his trustworthiness.”

With all the data that digital media affords, there is a danger that we can get caught up in the numbers, measuring people and their behaviour is never going to be an exact science. Whilst we can take the data available as part of the picture we must remember that there are many external factors that can affect influence. People are complex things and we must be careful to place a numerical figure on behaviour implying causation.

With that in mind, below is a simplified model which can help as a framework along with the data available for examining a situation of influence. It is hard to project forward many of the factors at play but can help along with the data to put together a picture.

Influence:

Any situation of influence involves 2 parties, influencer & target.

1. Influencer’s ability to influence relies on:

Credibility (what they know): The influencer’s expertise or perceived expertise by the target on the subject matter.

Social Capital (who they know): What standing does the influencer have in the community, do they know and associate with people that support the area of influence in consideration.

Bandwidth: The extent of the influencer’s ability to transmit their knowledge through their network of channels.

2. The target’s likelihood to be influenced by a specific influencer:

Relevance (the right information): To what extent do the target’s information needs coincide with the influencer’s expertise and subject matter.

Timing (the right time): The ability of the influencer to deliver the information to the target at the time when the target needed it.

Alignment (the right place): The extent of the channel overlap between the target and the influencer across their networks. If the two parties overlap on many channels, the expectation would be an increased situation of influence.

Confidence (the right person): How much the target trusts the influencer with respect to their information needs. This comes back to the influencers credibility and social capital in the eyes of the target.

Alex