I had a skip in my step as I walked the 19 blocks to the electronics store. It would be a much slower, disappointing walk home — and one that would change my relationship with this retailer forever.
I had decided that it was time to get a true wireless headset. I had done the research and knew exactly what I wanted. But wanting to support my local stores, I went to this major retailer’s website to see if they had it in stock locally. Their website said they had several in stock, so I headed off, excited to get my new electronic goodie.
You know where this is going.
When I got there, there were no headsets to be found. Their system was not updated in real-time — leaving me with incorrect information and a wasted afternoon.
I was dejected and frustrated. And have not purchased a single item from this retailer in the three years since.
Today, customers — and, for that matter, employees — expect everything to happen in real-time. Beyond that, everything from ever-decreasing switching costs to the competition for talent is increasing the risks and rewards of an organization being able to react in real-time to continually changing data.
Like it or not, we’re now living in a real-time world. The question is: how will this impact your organization, and what can you do to respond?
Why Real-Time is the New Black
In a recent white paper, Jason Bloomberg examines this broad question. And to answer it, we need to begin with breaking down just how challenging it is to create what he calls real-time performance.
Real-time performance is the idea that organizations now need to make critical business decisions and feed data and insights into customer interactions at the point and time the data is generated. It’s taking what sounds like esoteric technical terms (e.g., real-time computing, data streaming, etc.) and puts them in a business and customer context.
Put simply, what organizations need is the ability to execute and perform in real-time to meet the needs and expectations of their customers and employees. But while that intuitively makes sense, particularly in today’s modern always-on context, it is incredibly difficult to deliver from a technical perspective.
To begin with, there is a tsunami of data arriving from numerous sources and transactions. Organizations need this data to meet expectations, and humans need to be able to understand it, make decisions based upon it, and take actions accordingly — no small feat.
Second, meeting customer and employee expectations demands more than transactional data accumulated into batches and then processed.
It requires context, which means data that almost certainly comes from multiple sources and in various time dimensions. Putting that all together is a heavy lift and, as Bloomberg points out, begins to run into the laws of physics as connecting to, accessing, and leveraging all of that data — especially when it could be sitting in the cloud anywhere in the world — takes time and adds precious moments to an interaction.
And those are moments that may very well determine the customers’ or employees’ perception of the experience.
Today’s real-time expectations are pushing traditional technology architectures past their point of performance. To compete now and in the future, organizations are going to need to approach the challenge differently.
How the Dynamic Duo of In-Memory Data Stores and Stream Processing Can Change the Competitive Landscape
As the world becomes more focused on real-time engagement, organizations need to understand the essential components that can enable them.
As Bloomberg explains in his white paper, several elements come together to help organizations deliver real-time performance. Of these, however, it is the combination of two of them that, when combined, can create a definitive competitive advantage: in-memory data stores and event stream processing.
“Hazelcast’s true secret sauce, however, is the combination of its in-memory data store and real-time streaming compute capabilities – a combination that delivers on real-time performance requirements across a wide variety of enterprise use cases,” says Bloomberg. “Without Hazelcast, real-time intelligence tends to run afoul of one bottleneck or another. Hazelcast’s architecture resolves such bottlenecks, removing limitations to business performance, opening up new possibilities for technology capabilities, and in turn, business value.”
In-memory data stores help overcome part of the problem by enabling the data architecture to eliminate bottlenecks that may slow performance, such as submitting the request over a network connection and waiting for a disk to find and deliver the requested data. Essentially, it enables real-time performance by bringing data to the point of compute and eliminating the delays that traditional distributed computing wrought.
But that’s still only half of the equation.
Having the data accessible with very low latency is part of the solution. However, you still must be able to process that data as it becomes available in real-time to deliver real-time performance — consuming and processing it in batches is insufficient because it means that it won’t be usable for minutes, hours, or days after the data was generated — a lifetime in the real-time economy.
That’s where the combination of a real-time stream processing engine, combined with an in-memory data store, starts changing the game.
The Intellyx Take: Don’t Underestimate the Power of Real-Time or Get Scared Off by the Techie-ness
While all of this may sound geeky and so much tech-speak — and something you should leave to your technical team — it’s essential that you understand the impact of these types of technologies on your ability to compete in the future.
While you don’t need to become an expert in these proven, enterprise-grade technologies, you must recognize the challenges that traditional, legacy architectures you have in place will encounter as greater real-time demands are placed upon them. In most cases, they will simply be unable to meet anything near real-time customer and employee expectations — and their desire for personalized experiences.
And while there are many facets to creating personalized and differentiated experiences for your customers and employees, almost all of them — at least eventually — will run smack into the challenge of not being able to deliver the data and insights fast enough to make the right decisions or deliver the right type of interaction.
Understanding — and putting into production — the technical components that underpin your ability to deliver real-time performance, therefore, is important as you craft your go-forward strategy. And it’s this need to deploy real-time performance capabilities that is leading organizations to turn to intelligent application platforms, like Hazelcast, that combine in-memory data stores with stream-based computing.
A critical first step in your educational process is learning more about these technologies and the specific role these new types of application platforms can play within your technical architecture. You can begin that process by checking out Bloomberg’s white paper, Two Dimensions of the Real-Time Intelligent Application Platform.
But whatever you do, do not underestimate the power of real-time nor get scared off by the technie-ness of what enables it. They will be the key to your competitiveness.
Copyright © Intellyx LLC. Hazelcast is an Intellyx client. None of the other companies mentioned in this article are Intellyx clients. Intellyx retains full editorial control over the content of this paper. Image credit: Andre Hunter.
