From start-ups to Fortune 500 organizations, we’ve observed the following trends emerging in big data analytics that have shaped our areas of focus for 2015.
Make the right investment, soonerA common question we’ve observed from clients is: “We have lots of business ideas but how do we choose the right idea to invest in?”
One of the traps of working with new ideas and lots of data is the temptation to collect, cleanse and transform data before getting started. This can sometimes takes months to years without a single insight gleaned. Starting with the principle that ideas can be converted into a series of time-boxed experiments, we apply qualitative research techniques to test early hypotheses about the opportunity. Ideas are then validated by the data which provides data scientists with a direction to pursue before even looking at the data.
Get access to your data, fasterOvercoming the daunting task of getting started with predictive analytics requires access to “raw” data to assist with experimentation and proof of concepts. Techniques such as the Data Lake and Lambda Architectures enable teams to test out hypotheses by delaying the high cost and burden of up-front data modeling, MDM, data governance, cleansing and transformation, and in turn focuses teams on building the right thing, faster.
These techniques are gaining traction and you can read more about it in the December edition of ThoughtWorks Technology Radar, a bi-yearly publication that examines technology trends.
Stay relevant in a fast changing industryR or Python? Tableau or Qlikview? Hadoop or Spark? We are seeing lots of technology leaders struggle with which technologies to embrace, which vendors to believe, and how to develop the in-house skills to deal with these new technologies.
Staying up-to-the-minute on every new technology can be exhausting, and sorting through vendor hype is very difficult. Good solutions aren’t about the technology. Instead they are about aligning business goals with a value-driven approach, and selecting appropriate technologies. Appropriate technologies need to be cost effective, well-suited to the broader IT infrastructure, sustainable, scalable, and designed for the job at hand.
Strike the right balance between data and privacyDeveloping a trusted partnership between a consumer and a provider is tricky. Consumers have increasingly high expectations about how their data will be used, whom it will be shared with, and how it will be protected. Retailers and other providers violate that trust at their own peril.
Four potential dimensions of influence over customer data privacy include:
- Governmental policy and compliance,
- A practitioner code of ethics,
- Corporate data collector responsibility, and
- Consumer personal awareness and attention.
It’s an exciting time for big data analytics. The growing plethora of big data technology choices at our fingertips has enabled a new landscape to create disruptive new business models and engage with consumers in ways never experienced before. Looking forward to 2015, the complex balance between trust, privacy and corporate responsibility presents some interesting challenges ahead.