Não perca: Inspire 2024, de 13 a 16 de maio de 2024, no The Venetian em Las Vegas! Inscreva-se agora mesmo.


Why Business Executives Should Back a Data Lake Investment

Strategy   |   Bertrand Cariou   |   Jun 2, 2016

Let’s be honest—the tech industry has gotten really good at creating new concepts and over-promising their results. “Big data” and “data lakes” have been trending buzzwords for the past five years, but is there any evidence of benefits that justify its adoption and the increase in data lake analytics? Also, what is a data lake?

To start, it’s clear that business leaders rely on data.It’s essential in defining internal strategy, such as team objectives and metrics, as well as externally-facing ecosystems, such as the relationships between the business and suppliers or revenue drivers (customers, claims, tickets, etc.) Business leaders know they can’t manage what they can’t measure. With data, it’s always a balance of efficiency and innovation, of leveraging existing data to optimize current operations, while also looking ahead at new data sources, including the cloud-based options, and new initiatives.

But one thing’s certain: none of this data is carved in stone. Driving a business forward means maintaining a constant feedback loop of execution and results to continually improve and adjust. And, since business models are variable by nature, digital investments (particularly big data) must support the need for dynamic analyzes. This is where data lakes come in—they play an essential role in supporting more agile data lake analytics to reflect a constantly evolving business.

Despite its name, the primary value of big data technology is not necessarily to manage large volumes of data. Below, we’ve listed some of the under-the-radar value drivers of big data, in reverse order of popularity.

5. Cost Efficiency: Mitigate the risk of investment.

What’s one of the underlying factors that prevent business executives from investing in big data technology? Cost. The fact is, buying any new expensive new technology when you’re not sure what you’ll get out of it is risky. 

Typically, price is always a good reason to invest in a new technology, but basing your technology decisions solely on cost reduction (redoing the same old stuff but with a new technology) doesn’t bring a lot of business benefits. There are better benefits to be found elsewhere to drive business agility.

4. Data Variety: Enrich analytics with untapped source of information.

Big data opens up a new world of data types to uncover deeper and broader insights for the business. While traditional analytics solutions can only manage what is typically referred to as “structured data,” such as transactional systems, big data allows businesses to analyze a variety of new data—Twitter tweets, comments on Facebook, webchats, web navigation information, or machine-generated data. Big data technology and data lake analytics enable easy storage and processing at scale of traditional data, as well as allowing businesses to leverage new types of data, and those to come.

3. Flexibility: Agile analytics at the speed of business.

Next time you want to impress your colleagues, explain that you’re adopting big data because it’s “schema on read.” What’s so impressive about that? While the traditional approach to analytics mandates that users decide the structure (schema) ahead of time, big data technologies allow users to structure it on-demand (the schema is created on-read = schema on-read!). 

You no longer need to know ahead of time all possible questions you may ask – you can interrogate the data whenever and however you like based on your upcoming requirements.

You can add new data sources anytime. You can innovate faster.

You can create reports with data lake analytics in hours instead of months – so you can measure what counts at the speed of your business.

2. Data Lake: One place for the supply and demand of data.

This is a huge reason for adopting big data, and more specifically, a data lake pattern and data lake concept. Every department generates data that theoretically could be used across the organization. But without a data lake and efficient data lake architecture, this data stays in silo. This data fragmentation holds back innovation, or simply prevent in responding to a business question efficiently.

An enterprise data lake stops the data fragmentation headache by bringing all the data together in one place so users can, thanks to the flexibility, respond more easily to a question. Practically, a data lake and data lake analytics help in bringing a consolidated view of customers, or products usage. Even if some data hasn’t seen much usage, it can be ingested into the data lake architecture for future use (thanks to the low cost of storage and processing).

1.Self-Service: Develop use cases as you go.

Business units often depend way too heavily on IT. Not only does this drain IT’s time and resources, but their schedules are often measured in years, even decades, while business units work on initiatives with a much shorter time frame. There’s a need to remove this friction between the business and IT.

The Data Lake, other data warehouses, and all the previous listed benefits would not be of any value without a self-service approach to using this new technology. Business needs to play with the datato deliver insight. Unfortunately, manipulating Big Data is not an easy endeavor, even for software engineers (you have to learn very strange languages such as Pig, Mapreduce, Spark and other esoteric names that pop up every day.)

This is where Designer Cloud shines, hiding the complexity of big data by providing a self-service approach for business users to prepare data for their analysis and data lake analytics. Without data preparation, the lake may freeze with a thick shape of ice preventing access to this wealth of information (unless you like ice-skating, your lake will be of no use). Data preparation, more often known as data wrangling, is an intuitive approach to achieve data manipulation. Designer Cloud is just as easy as Excel to explore, combine, blend, clean and de-dupe the data before publishing it for your reports or next phase of your analytical process, with simple integration of current data analysis tools. It’s a critical component that will make a Data Lake strategy and data lake analytics successful to align with the requirements of using the data assets at the speed of the business.