Behind the Buzzword: The Reality of Real Time

Imagine Uber or DoorDash working without real-time updates. How would you know when to look out for your ride or food? We’ve been programmed to think faster is better. This mindset has made its way from our consumer brand relationships to Silicon Valley and B2B tech as well. Customers who have bought into this concept assume that data solutions with zero latency are the best answer to their data management needs. While someone monitoring fluctuations on the stock market may gain a competitive advantage from a microsecond response time, in many industries, too many updates can create unhelpful distractions.

Real time can mean many things. For example, when it comes to data, real time can mean once a day, in an hour, or even a microsecond. To strategically control and analyze data, companies need to know which version of real time works for them.

To do this, it’s important to align your business goals. Start by understanding the frequency of action and inspection and then matching your technology to that. In some cases, you will need to shave seconds off to gain competitive advantage, but the reality is that most situations don’t require that fast of a response. Customer lifetime value is not changing every minute, so measuring it second by second can give you a false sense of precision.

One example of this is Peloton. Peloton delivers real-time cycling experiences to its riders via video. To maintain engagement and motivation throughout the workout, Peloton needs to deliver high-quality and consistently streamed video, so low latency is critical. However, this is easier said than done when you have thousands of riders simultaneously tuning into one class. To deal with these latency ups and downs, Peloton uses Datadog and AWS to monitor performance and adjust accordingly.

Similarly, Paddy Power Betfair (PBB) has to track betting and game results down to the second in order to deliver the best customer experience. But, the sports betting platform needs to manage massive volumes of data, access and serve it with incredibly low latency levels to do this. What's more, PBB’s data volumes are not only massive, but they vary tremendously over time. To manage these spikes, PBB leverages the cloud with a data platform on AWS S3, Redshift and Aurora.

Both Peloton and PBB have different definitions of real time and different levels of latency to match. Customers expect different experiences but all expect the companies to perform and deliver their data seamlessly.

Practical real time

The goal of real-time data is to decrease the time between an event and the actionable insight that could come from it. In order to make informed decisions, organizations should strive to make the time between insight and benefit as short as possible. The key to being agile is to minimize the friction that can occur. Doing so will make data more flexible and more adaptable.

Utilizing an integrated platform in the cloud can cut down on the latency that legacy infrastructure models offer and increase time to value. Latency in processing occurs in traditional storage models that move slowly when retrieving data. Organizations can decrease processing time by moving away from those slow hard disks and relational databases, into in-memory computing software. However, research shows that every 100-millisecond delay in website load time can hurt conversion rates by 7%. To mitigate these issues, it’s important to establish a baseline performance to monitor against, and then identify the causes of performance degradation. Depending on the network -- and the latency that comes with it -- “real time” can vary.

When determining your reality of real time, begin with the end in mind and think about the “why.” Data is always being collected, transformed and analyzed, but what’s important for real time is what you’re going to do with the data report. Think of how often you need to report your data to the end consumer of that data, and the cadence should follow suit -- for example, reporting to the board, executing a financial transaction or building a customer profile.

Real time should not be based on what the end consumers say they need, however, because they will always want it faster. Be realistic. Ask your stakeholders what they want, why they want it, and what action they’re going to take from it, and use that as a guide. Real time comes with tradeoffs of cost vs. speed, so work within your budget at the acceptable level of latency to your audience.

Real time is different for every company and every industry, so identify what works best for your organization. Don’t get caught up in the hype. Understand what both you and your consumers’ priorities are and work backwards. The reality is, there is no such thing as “real time.”  Move beyond the buzzword and deliver the best result that works for you.

Jake joined Talend following the acquisition of Stitch, where he was co-founder and CEO. In his current role of senior vice president at Talend, Jake oversees Stitch as well as Talend's other frictionless go-to-market initiatives. Prior to Stitch, he co-founded RJMetrics, which went onto be acquired by Magento in July 2016. Jake graduated from Wharton with high honors and a degree in entrepreneurship and finance.