How can organisations use data effectively, according to Microsoft Azure

The largest enterprises have amassed an incredible amount of data from a variety of sources. But, the problem is identifying that data, accessing it and using that most precious asset effectively.

Cloud providers like Google Cloud, AWS and Microsoft Azure can provide tailored services that help organisations break down these silos and derive real value from what many have coined the new oil.

In that vein, Information Age spoke to Rohan Kumar, corporate VP of Azure Data, at Big Data LDN about how organisations can use their data effectively.

The data problem

“One of the biggest challenges we see from our enterprise customers surrounds the core enterprise data warehouse, which effectively contains the crown jewels of data’ customer information, employee resource information etcetera,” began Kumar.

There are a number of problems enterprises face in utilising this hoard of data, such as building end-to-end solutions that need to ingest data, integrate data, which eventually leads to data transformation. It’s a very complicated, costly and timely process that requires organisations — independent of the public cloud — to stitch together multiple services, such as maintaining the pipeline and making sure security up to standard.

It’s the same problem with machine learning where the biggest challenges are not around the code algorithms itself, but rather, DevOps. Users are struggling to track all their experiments they had on the right datasets and eventually pick the best one to put into production.

Competitive advantage: how to put machine learning models into production

Most CTOs highlight machine learning as the technology that will disrupt their industry and lead to new innovations. But, how can they effectively put machine learning models into production? Read here

Using data effectively

However, this whole arena is getting disrupted in a big way and it comes down to two fundamental reasons; organisations want to:

1. Scale — with the rise of big data, click streams from e-commerce sites, connected cars, the list goes on, there is a challenge of scalability. It’s important to be able to use data to drive projects out on a large scale.

2. Moving from traditional to predictive — on top of that, there is a move from traditional analytics to predictive analytics, which uses machine learning to do very advanced statistical analysis to predict the future.

A solution: one unified product for analytics

Microsoft Azure recently announced a solution to this, which it says can help organisation effectively use their data.

“For us Synapse Analytics was about merging both these worlds, the traditional world of warehousing with the more predictive world of big data, get that into a single system and work that at scale to solve some of the big customer problems we see,” said Kumar.

“We believe Synapse is a category defining product that looks at multiple products and services and puts them in one unified product experience.”

Nallan Sriraman, global head of technology at Unilever, an Azure customer, confirms this praise from Kumar, stating: “Our adoption of the Azure Analytics platform has revolutionised our ability to deliver insights to the business. We are very excited that Azure Synapse Analytics will streamline our analytics processes even further with the seamless integration the way all the pieces have come together so well.”

Defining success

How does Azure define success for itself and its customers?

In the line of business almost every customer wants to move quickly. It’s frustrating when users know the question they want to ask, but can’t get an answer immediately, because data is so disparate.

Finding data quickly — how can this Silicon Valley AI startup help your business stay ahead of disruption?

The largest enterprises need help avoiding disruption from agile, digital native challengers. These Silicon Valley AI startups can help address your greatest pain points. Up second, Promethium. Read here

“Business leaders need the answers to their questions today, but it can take technical teams months to find the answer,” explained Kumar.

“This problem became very apparent when large customers joined Azure,” he continued. “Our goal, then, for measuring success, was to help our customer run experiments, gather data and answer business critical questions quickly. The aim was to reduce the query time from several months to a day. That’s what success is to us and we believe we have done that with Synapse, but the proof is in the pudding.”

The aim of this solution is to help organisations work in a more agile way and for users to get the insights they need to run their business, very quickly.

The pudding: customer quotes

“We believe that Azure Synapse Analytics can provide the scalability that an organisation like Deutsche Bank requires for its analytic workloads, while reducing the complexity of our infrastructure. This will allow us to focus on our data, and value we can derive from it” — Paul Russell, head of Engineering, Enterprise Analytics Platforms, Deutsche Bank.

“In the modern data landscape, Azure Synapse Analytics is the missing link that enables discovery on massive data volumes, quickly, easily and with the minimum of fuss, maximising value for our business” — Grant Nairn, CIO, Aggreko.

“We anticipate Azure Synapse Analytics will drive faster insights from all our data, and thus, a faster path to better decisions and improved returns” — Danny Siegel, vice president, Information Delivery, Newell Brands.

Reduce complexity

Data analytics and AI have become a significant part of empowering every human and every organisation to achieve more.

But, the benefits of these technologies and the value they can produce won’t reach their potential, unless the complexity around them is reduced.

“How much effort goes into building things out to achieve insights is not sustainable and currently we won’t achieve that data-driven culture we desire,” said Kumar.

The complexity of analytics and machine learning is not easy or commonplace. The industry needs to make it simple for the data engineering population to use the power of machine learning with the toolsets they know. Expecting everyone to raise the bar and become a data scientist is not practical.”

Avatar photo

Nick Ismail

Nick Ismail is a former editor for Information Age (from 2018 to 2022) before moving on to become Global Head of Brand Journalism at HCLTech. He has a particular interest in smart technologies, AI and...