2016 Trends in data

Publish date:

Seen any good movies lately? This certainly seems to be the time for it. The new Star Wars. The new Tarantino. It is also a good time for some 2016 trends and predictions. Movie style, with some pretty catchy quotes, to keep in fashion. In random order: 1. “Catch Me If You Can” – Customer […]

Seen any good movies lately? This certainly seems to be the time for it. The new Star Wars. The new Tarantino. It is also a good time for some 2016 trends and predictions. Movie style, with some pretty catchy quotes, to keep in fashion. In random order:

1. “Catch Me If You Can”Customer and operational analytics get real, real time

Customers – and citizens alike – expect easier access to online services, resulting into increased digital interactions. In order to recognize and serve customers with the right set of products and services, customer analytics are essential. And they need to be available on the spot, as the moment of connection may be gone before you know it. Also, in order to stay in sync with the external rhythm, analytics should be used in exactly the same intensity to optimize operational processes. In 2016, we will see the rise of powerful platforms such as Spark and Nifi, bringing the in-memory power and the development capabilities for building real, real time analytical solutions.

2. “Back off man, I’m a scientist”DIY data science hits the executive floor

The DIY way – that we already have seen for some time now in Business Intelligence and reporting – will now quickly spread into the areas of analytics and data science. Tools such as IBM’s Watson Analytics and BigML enable potentially any business executive to be a bit of a data scientist, exploring algorithms and predictive models all by themselves. In order to make sense of the abundance of data, visualization will be key to see the forest from the trees. DIY data science may be instrumental to creating a true data-driven culture within the business, but amateurism – even of the enthusiastic kind – may result into wrong insight.

3. “Use the Force, Luke” – Insights can be driven by crowd power

It is tough enough to set up and maintain a big data platform as it requires new, scarce skills – like data science – and new, unexplored technologies – like Spark and Hadoop. It’s also difficult to project when, how much and how often you will need these new, purple unicorn capabilities. So why not look outside and get social right from the start? Open community platforms such as Kaggle provide organizations with the flexible, collective brainpower of some of the smartest data scientists and technologists in the world. Also, have a look at the quickly emerging market of catalog-based, off-the-shelf analytics as a much quicker way to get going on your insight-driven journey.

4. “Let It Go”Machine intelligence starts its own journey

Machine intelligence changes the way enterprises gain insight from their data. It makes use of deep learning, pattern recognition, cognitive reasoning and complexity reduction to create analytical models that often surpass what simple mortal souls can absorb or create. It potentially reduces human intervention and can even fully automate the building of new analytical models. Organizations can thus highly optimize their solutions in areas such as fraud detection, asset management or product recommendations. But the potential of autonomics, in which independent machine intelligence truly augments what humans do, eventually goes far beyond that.

5. “You Talkin’ To Me?” – IoT goes both ways

Yes, definitely yes: we’re still getting more and more physical: The Internet of Things (IOT) will continue to evolve as one of the strongest drivers for the next wave of ‘Big Data’ analytics. And it will not be limited to ingesting huge volumes of data from sensors for later analysis in a central environment. In order to reap the real benefits of IOT, insights should be pushed back to things – often in real time -as well. This does not only pertain to the personal IOT (e.g. direct feedback on fitness, weight, sleep, activity), but also to the industrial IOT. There, machines and sensors are connected with insights, leveraging for example predictive maintenance and usage optimization in high-stake industries such as energy, manufacturing, automotive and chemicals.

6. “You’re Going To Need A Bigger Boat”The data landscape evolves in different ways

With all the potential and promises around Big Data, organizations need to redesign their data landscape in order to deal with the volumes, timing considerations, agility needs and analytical requirements that come with it. Many consider to offload parts of their current – expensive – data warehouses to low-cost/mass volume platforms such as Hadoop. They also look into a new breed of analytical products – such as Tableau and R – that may be different from what they have been using so far for enterprise reporting. The data architect in 2016 needs to be able to seamlessly blend various flavors of existing and new technologies, keeping in mind that a highly distributed, federated data landscape will be more and more the typical enterprise reality.

7. “Why So Serious?” – Towards the next maturity level

Big Data needs to grow up, if it likes it or not. The early, exciting days have been very much driven by agile, exploratory approaches and rogue Proof of Concepts across the enterprise space. It helped gain momentum. But now, the next maturity level needs to kick in. Definitions need to be aligned, quality needs to be ensured, production needs to be industrialized, technology standards (such as the ODPi) need to be agreed on. A company-wide data-driven culture – including agile Do It Yourself analytics – requires an enterprise-scale foundation. Think about topics such as data classification, data sharing, privacy, local regulations, security and access control to start with. To finally go beyond that infamous Proof Of Concept, it’s all a matter of getting serious.

The end

Did your recognize all quotes? Maybe you have come up with some great suggestions yourself? Do not hesitate to share them. In summary, we predict that enterprises will create more insights themselves – with the risk of ridiculous analysis backed up by heaps of data –  or get it as a service – with the risk of placing their faith in other mans hands. There will eventually be a recognition that they must start to define where data has come from, its history, its accuracy, the confidence they have about it.  Data & Insights deserve an integral approach. In other words, you need to understand the data before you can understand the insights.  Only when the basics are fixed, enterprises can spend less time on data in order to have more focus on the business outcomes of insights, in creating value. Remember, in the end, it is about the beginning.

PS: This blog could have not been created without the help and support from my colleagues from the Global I&D Architecture & Advisory practice: Cyriel Houben, Rüdiger Eberlein, Sanjay Khangarot, Shivprasad Rao, Hemal Vora, Arnout Arntz, Satyajit Mohanty, Stephen Timbers and Ron Tolido.




Related Posts

big data

Moving Big Data delivery from the West Coast to the East Coast—part 1

Simon Turnbull
Date icon January 17, 2018

Over the past few years we have started to see a paradigm shift in the capabilities that...

Business Information Management

SAP and Today’s Business to Business Transaction

Kenneth Van Meter
Date icon April 25, 2017

When companies begin to think about how they can reach their prospects, they often turn to...

Business Information Management

How Should a Tester Adapt to Cloud – Call for Change of Mindset Amongst Testers

Date icon January 23, 2017

It’s time that we as testers prepare ourselves to master and test this inevitable...