Tag Archives: social analytics

How Sprint Used Digitas’ BrandLIVE During CES

Beet.tv interviewed Anne Marie Kline, of Digitas, at last week’s CES show. In a short video clip, she talked about how Digitas used its new BrandLIVE platform to track conversations online that were related to topics that Sprint was interested in, and then create a response in real-time, or near real-time.

I wrote about BrandLIVE when it launched late last year. 


Walmart Labs on the Challenges of Real-Time Social Analytics

Arun Prasath, an engineer at Walmart Labs, describes the retailer’s 2013 social analytics agenda in a recent blog post. What he writes crystallizes the excitement—and terror—of confronting the idea of using social data in real time:

Social data mining comes with incredible challenges, which only makes it all the more exciting for our super smart engineers to come to work every day. Data volume is formidably huge. We are talking about petabytes here. Real-time social data processing requires sophisticated data stores and blazingly fast algorithms. The noise levels are exorbitant, the language used in social forums is heavily informal, unstructured and often ungrammatical, and filtering out that helpful insight out of the huge amount of noise is super hard. Just consider algorithmically parsing – “OMG!!! dis is sooo coool! i luv ma new fone. i cant believ ma luck 4 chosin this! #wellwhatdoyathink”. Popular text analytics and natural language processing techniques based on standard language models simply fail. We need altogether different techniques to filter out and focus on social data that is relevant to us, which in itself is a daunting task. The next step is to map this to meaningful retail products. All of these are difficult tasks.

Real-Time Marketing: The Questions I’m Pondering

I’m getting caught up on posting things to this blog … the holidays sucked away all my time and then I got deep into researching a report. I’m looking at how marketers are using social analytics to support their real-time marketing efforts, and how the analytics vendors are evolving to help them. It’ll be my first foray into covering real-time marketing as a report topic and I’m excited, if also a bit overwhelmed by all there is to think about. Turns out real-time marketing is a topic with a lot of moving parts to it.

Here are some of the questions I’m pondering, not just for this report but on the larger topic of real-time marketing and what it means:

How real is real time? Is it milliseconds, minutes, hours, same day, tomorrow? An ad that is targeted on the fly in milliseconds is considered “real time,” but so is the idea of latching onto a trending topic an hour or two after you notice it’s trending. Is real time defined differently depending on your marketing goal?

Is there a role for humans or is real-time marketing completely an automated function? Yes, humans write the code and run the analytics, but to what extent humans should be involved is interesting to ponder. Humans have a unique capability to steer machines away from making a misguided decision (such as latching onto the wrong trending topic).

What role does social data play in real-time marketing? Primary role? Secondary? Essential? Nice to have but not necessary? Companies are collecting myriad datapoints in their efforts to get more responsive. Social data is enormous in quantity, difficult to interpret and extremely ephemeral. Plus it’s hard to know if the insights you are getting from social media are real or not. So how much should marketers really rely on social data as they plan out their real-time marketing efforts?

Are marketers prepared to do real-time marketing yet? I’m guessing that on an experimental basis, yes, but as a regular part of their marketing plan, no. Most of what I see are one-off campaigns. There’s social CRM, of course, but while that activity is real-time in the sense that marketers try to respond as quickly as possible, it seems to fall more in the category of customer care and less in outbound marketing.

So, what questions do you have?

Forrester Report Helps Marketers Understand Social Intelligence

Forrester recently published a report called “Ten Steps to Successful Social Intelligence Measurement,” aimed at helping those who work in customer intelligence to better use social signals in their analytics. CMS Wire describes the 10 steps, which include “separate reactive social modeling from proactive social intelligence”. The reactive stuff is the domain of PR, CMS Wire says, while proactive work is the domain of customer intelligence. Forrester also suggests that marketers should “get help from integrated technology systems” – because “successful social intelligence requires real-time data capture, analysis and action.”

New Digitas Unit Gets Into the “Now” of Marketing

Digitas is embracing social monitoring and the content marketing explosion with a new unit dubbed BrandLIVE. The unit is designed to provide such services as social listening, real-time analytics, content-sharing measurement, “proactive and reactive content,” and a publishing/distribution system that enables real-time tracking and optimization.

The unit will be led by Anne-Marie Kline, Digitas senior VP-marketing, and John Robinson, senior VP-creative. In announcing the unit, Kline told Adotas:

“BrandLIVE is a deft intelligence center … that conditions brands to respond to the unexpected and keeps that brand in the now with an eye on the future.”

Startup Jetlore Enables Real-Time Content Decisions

TechCrunch covers a startup called Jetlore today. The company mines social data and applies context to deliver its analytics. Key phrase in this article: Companies can “use that data to make content decisions — if a lot of our users are tweeting and commenting on a given tech-related topic, we probably want to write a post addressing it.”