Updates from November, 2015 Toggle Comment Threads | Keyboard Shortcuts

  • admin 9:46 am on November 30, 2015 Permalink
    Tags: , , , , , ,   

    Teradata Partners How to manage data from the Internet of Things 

    Teradata Press Mentions

  • admin 9:52 am on November 29, 2015 Permalink
    Tags: , ,   

    What to Avoid in Customer Analytics 

    Working recently on enabling customer analytics for a large Australian bank, I was once again reminded of the potential of Big Data and the risks involved.

    This bank proudly analyses customer transactions to glean important information about its customers. They plan to use the data, for example, to offer you travel insurance when they see that you bought a flight ticket. But they also plan to offer you access to their lounge in Singapore if the flight tickets were bought from Singapore Airlines.

    The big question is: as a customer, would you be happy to receive such an offer? Where does the (very subtle) creepiness line pass? How do we decide what’s acceptable and what’s not? Some examples might help.

    Let’s start from the creepiest. I know of a US-based company that offers software that analyses web-clicks and can predict with high precision, for female surfers, whether they are having their period. Creepiest, you must agree. Not surprisingly, the company doesn’t get a lot of clients (no female clients, I assume).

    A more famous example is the Target story, where their analytics predicted a young girl’s pregnancy before her family knew. What did Target do when the story came to light? “We found out that as long as a pregnant woman thinks she hasn’t been spied on, she’ll use the coupons. She just assumes that everyone else on her block got the same mailer for diapers and cribs. As long as we don’t spook her, it works,” —said Target to Forbes.

    On the other side of the continuum there are companies who “spy” on me and make it worth my while. Two examples are Google and Amazon. When I type a search item, Google attempts to complete it for me. It uses what I searched for in the past and what you are searching right now. It spies on all of us and we love it. Why? Because the benefits outweigh the loss of privacy.

    We rarely feel that our privacy has been compromised by this and we enjoy the benefits.

    Similarly, when I buy a book from Amazon, they tell me that “people who bought this book also bought those books…”. Again, they are spying on me and you, creating a detailed profile of my buying habits and comparing it to your profile. Do I feel that my privacy is breached ? Not at all. The recommendations are actually quite good, usually. Does this benefit Amazon? You bet it does. Fortune magazine claims that Amazon get up to 60% conversion rate on their recommendations.

    A middle-way example is Orbitz. As a result of customer-spending analytics, Orbitz decided to present Mac users with more expensive options than PC users. All users had access to the same offers, but Mac users would have to work harder to see the cheaper options. As an Orbitz user, would you be happy with this? Would you complain? Or would you switch to a PC?

    Your company is likely to find itself somewhere between these extremes. Yes, you want to know as much as possible about your customers. But you must ensure that you don’t alienate them by compromising their sense of privacy.

    My advice: do your customer analytics, but use the results wisely.

    Ben Bor is a Senior Solutions Architect at Teradata ANZ, specialist in maximising the value of enterprise data. He gained international experience on projects in Europe, America, Asia and Australia. Ben has over 30 years’ experience in the IT industry. Prior to joining Teradata, Ben worked for international consultancies for about 15 years and for international banks before that. Connect with Ben Bor via Linkedin.



    The post What to Avoid in Customer Analytics appeared first on International Blog.

    Teradata Blogs Feed

  • admin 9:48 am on November 29, 2015 Permalink
    Tags: , , , , , , tricky, Turning   

    Turning data into action still tricky for Australia s data scientists 

    Teradata Press Mentions

  • admin 9:53 am on November 28, 2015 Permalink
    Tags: , Cyberspace,   

    Build Loyalty in Cyberspace 

    Q4-15_Theme_Social Media

    by John Edwards

    In cyberspace, every interaction can be broadcast to a person’s friends and contacts, which influences their views. In fact, losing a loyal customer who “unfollows” a brand on social media can have a ripple effect since about 10% of those who unfollow will tell their friends to do the same, and approximately 10% post a status update reflecting the change, according to Adweek SocialTimes.

    “One influential person can have a bad experience and share it with their following, wreaking havoc on a brand,” says John Lovett, senior partner at Analytics Demystified, a consulting firm. 

    New Paradigm for Influence

    As social media becomes fully integrated into daily life, organizations must engage customers through all available channels, including websites, online videos, mobile apps and digital signage. This engagement, across the mediums people are already using, helps build and maintain brand loyalty at a time when customers are more likely than ever to switch brands.

    “Service has extended beyond the call centers and physical stores to the Web and social media,” Lovett points out. “Companies at the forefront of this shift are spending time and money to train their employees on social media and to deliver experiences that exceed expectations.”

    Negative online reviews, meanwhile, can wreak havoc on loyalty. “The power of choice is stronger than ever and brands no longer wield the influence they once did,” Lovett explains. “In today’s world, complete strangers who took just moments to review a product or service can have more sway than millions of marketing dollars spent on advertising.”

    Put Social Media Insights to Work

    Key insights obtained from social media sources, including Facebook, Twitter, Yelp, YouTube and similar services, can be used to better understand customer motivations, preferences and opinions of specific products. Companies can leverage these insights to deliver a positive experience. The information can also be used to identify problems with products or customer service, highlight new uses for products or point toward trends that could prompt organizations to add new features.

    John Edwards has covered the technology industry for more than two decades. 

    This article originally appeared in the Q4 2015 issue of Teradata Magazine. For more tips and best practices on strengthening customer loyalty in a constantly changing world, visit TeradataMagazine.com.



    The post Build Loyalty in Cyberspace appeared first on Magazine Blog.

    Teradata Blogs Feed

  • admin 9:47 am on November 28, 2015 Permalink
    Tags: , , Cuts, , , Problem, , , , ,   

    Turkcell cuts call problem resolution times in half with Teradata Data Warehouse Appliance 

    Teradata Press Mentions

  • admin 9:56 am on November 27, 2015 Permalink
    Tags: Bubble, Bursting, ,   

    Bursting The Internet Of Things Bubble 

    So Internet of Things (IoT) makes its appearance at the top of the Gartner Hype Cycle. Remember when that happened with big data? Well, as can be expected, IT vendors, magazine editors, academics and professional “experts” are all over the IoT bandwagon.

    Just like big data, you might not be surprised to learn that IoT has much of the same messaging, hype and buzzwords as big data (so no need to change too many slides or think too deeply about the messaging) and it is also “truly game-changing / a paradigm shift” (delete as appropriate). What’s not to like! Certainly if you are on the purchasing side of the desk you can look forward to a steady stream of salespeople calling on you with their “Internet of Things” solutions / conference / technology / revolutionary capability (take your pick!).


    Where’s the Value?

    Just as many companies have struggled to show value from their big data investments, I believe this will also be the case with IoT projects. When Jason Waxman, general manager of Intel’s Cloud Platforms Group, pointed out that “the dirty little secret about big data is no one actually knows what to do with it”, he could just as easily been talking about IoT.

    Early adopters of IoT think they know what to do with it. But when it comes to actually knowing how they will use it, how they can profit from it, that’s a little harder to do. And that’s because they have not answered the critical question – what business challenge am I trying solve?

    Theory of Evolution

    So why are so many companies struggling to answer this question? Ever since computers and data management were first invented there has been a constant evolution of capability, capacity and connectivity.

    This evolution is a key driver of IoT. As sensors become cheaper and more powerful they can be added to almost any device or machine. As connectivity becomes faster (and again cheaper) these devices can be connected to the internet, and in turn be connected to each other.

    From Tape to Data Lake

    What does this all mean? Take away the arguments about “hardware” and “interoperability” and what do you get from all these sensors and connected devices? Data. And yes, it is often big data. See above for how many companies are still struggling with making cents from their big data investments (pun intended) and you begin to question why companies are stuck in this infinity loop of paralysis, doing what they know/are comfortable with, and ultimately, disillusionment.

    In the 80’s and 90’s, leading companies understood the value of data and, in particular, of bringing together all the disparate data sources from across the company. They saw that, for example, by integrating production data, supplier data, service data and customer data, they could drive savings in warranty and service costs while also improving customer satisfaction and loyalty. By taking the data out of application and departmental strongholds, they broke down the silos between systems and between department functions.

    Now, it looks to me that while technology continues to evolve at a fantastic rate, the approach to data management and analytics in most corporations is regressing!

    I personally see a lot of companies building data lakes to store all their IoT data without any real understanding on how this is going to translate into delivering value back to the business. In many cases, they are just replacing tape with cheap storage! The net of which is that these are just new improved data silos – a clear oxymoron.

    The Whole is Greater than the Sum of All the Parts

    The truth is – IoT is a great enabler. But if we want it to revolutionise our business, then we can’t look at it in isolation. We need to combine it with data from across the organisation – customer records, usage rates, financial and supply chain logistics information.

    It is when we have the ability to analyse all the data, all of the time, through all of the processes, that we will be able to tackle complex business challenges – with confidence.

    Challenges such as, predicting when a piece of machinery might break down and taking it offline for service before that happens. Or knowing when to empty not just the Smart Bin on one street corner, but the bins from a whole neighbourhood.

    If IoT is to be the accelerant for your business, then choose evolution – get all your data together, use better and newer analytics, and prove your business case.

    This post first appeared on Forbes TeradataVoice on 29/10/2015.

    The post Bursting The Internet Of Things Bubble appeared first on International Blog.

    Teradata Blogs Feed

  • admin 9:46 am on November 27, 2015 Permalink
    Tags: , , ,   

    Teradata doubles down on IoT with two new tools 

    Teradata Press Mentions

  • admin 9:51 am on November 26, 2015 Permalink
    Tags: , , , , ,   

    Is Your Company Product or Service Centric? Do Your Customers Agree? 

    Mine their comments to find out.

    In an ideal world, all companies would have a portfolio of excellent products delivered with remarkable service. Competition in the marketplace however dictates that company resources have to be allocated where they can be most effective. Understanding whether customers see your company as product-centric or service-centric does therefore matter.

    Some companies have it easy: Apple for instance is (mostly) a product company. Airbnb is primarily a service company. But what about the retail side of Amazon (1)? What about banks, insurance companies, universities, or airlines? The line here is murkier, and all customers may not agree about the category. Moreover, ill-defined priorities increase the risk of a disassociation between the company internal view and its customers.

    For instance, a company can internally think of itself as product-driven (and communicate in that manner), while the majority of its customers are focused on its service side. The divergence of these views can generate disaffection among customers who don’t feel heard or taken care of by the company.

    Fortunately, there are steps one can take to understand what customers care about and align the strategic direction to the customers’ expectations. To do so requires stepping out of the ivory tower (academia is not alone in living in one) and listening to the voice of the customers.

    Directly asking people what they would like can be inaccurate because of the stated vs. revealed preference dichotomy often observed in consumer studies [1]. In addition, customers also view the world through their own prism, leading the famous Henry Ford quote: “If I had asked people what they wanted, they would have said faster horses” (Henry Ford probably never said these words, but they became memorable nevertheless).

    Reliable feedback can however be obtained from satisfaction studies, for instance when querying net promoter score (NPS). In that instance, customers are asked to give a numerical rating to a question such as “how likely are you to recommend company X”, and then are asked to justify/explain their rating. The collection of verbatim comments can then be analysed to provide valuable insight into perception customers have of the company.

    A great way to obtain actionable insights is to perform automatic topic discovery on the verbatim comments. Among the best performing methods to do so is Latent Dirichelet Allocation (LDA). LDA is an unsupervised machine learning model that automatically parses every comment and group similar ones together into a pre-defined number of clusters.


    Figure1: LDA example

    In the case of satisfaction studies, numerical ratings can be used as a pre-filter to understand topics customers are satisfied/dissatisfiedwith. For instance, with NPS studies, promoters and detractors can be separated prior to performing LDA (effectively we run 2 LDAs, one on promoter comments and one on detractor comments) to illustrate what customers like and dislike.

    As an example, we recently undertook the analysis of verbatim comments from NPS studies of a major client. We first separated the comments according to their promoter/detractor status (NPS score > 8 = promoter, NPS score < 7 = detractor) and performed LDA using Teradata Aster. The comments were clustered into 5 categories for both promoters and detractors. The key topics for each category is illustrated below:

    Promoter Detractor
    Personal Service All corporations are the same
    Never had a problem I don’t know enough, I am new
    Convenient, Easy Bad customer service
    Customer service polite and helpful Difficulties with products
    Good product experience Products too expensive


    One immediately sees that the quality of customer service and amount of customer effort (2) dominate the conversations, indicating that satisfaction (high or low) is primarily linked to service rather than products. As a result, service improvements will have a greater impact than changes in the product portfolio. There is also an opportunity for the considered institution to change its perception by showing to its customers it is not the same as its competitors.

    Verbatim comments offer a unique insight into users and customers due to being their own words. Carefully mined, these comments help define (or redefine) strategic priorities. Because focusing on products when customers care about service (or vice versa) will not make you successful in the long run.


    1 personally I view it mostly as service-centric

    2 In fact, customer effort score is probably a more predictive measure of loyalty and satisfaction than promoter score


    [1] “A comparison of revealed preference and stated preference models of travel behaviour”, M. Wardman, Journal of Transport Economics and Policy, January 1988

    Clément Fredembach is a data scientist with Teradata Australia and New Zealand Advance Analytics group. With a background in Colour Science, Computational Photography and Computer Vision, Clement has designed and built perceptual statistical experiments and models for the past 10 years.


    The post Is Your Company Product or Service Centric? Do Your Customers Agree? appeared first on International Blog.

    Teradata Blogs Feed

  • admin 9:48 am on November 26, 2015 Permalink
    Tags: , , Employs, , , , ,   

    From Farm to Fork McCain Employs a Data Driven BI Strategy 

    Teradata Web Casts

  • admin 9:48 am on November 26, 2015 Permalink
    Tags: , , , , ,   

    Teradata Looks To Supercharge Its Marketing Cloud With New DMP Acquisition 

    Teradata Press Mentions

Compose new post
Next post/Next comment
Previous post/Previous comment
Show/Hide comments
Go to top
Go to login
Show/Hide help
shift + esc