Application Development

The Future of Customer Experience Depends on the Digital Assembly Line—and Data

“Data gravity” has been a persistent drag on innovation across all organizations—and the power at our fingertips to solve this problem is making data programmable.

Alex Hesterberg

Oct 20, 2020

I am just over 100 days into my role at Delphix and I have been blown away by just how significantly data is playing a role in what some of us refer to as digital transformation or as I call it, less elegantly, “the biggest, most audacious platform shift in our recorded technology history.” The levers to design and realize technology tailored to consumer demand are now available to the broadest group of innovators across the globe. Whether you’re a start-up raising initial capital, an online car dealer, or a 30-year-old technology company, the opportunity to invent and reinvent is so prolific. It’s nothing short of astonishing.

The true inspiration for all of this incredible invention and reinvention capability are the applications that now, more than ever, represent the brand and customer experience for most companies today. This reality is ubiquitous—from car companies and hardware resellers, to grocery stores and banks. The digital experience is now simply the experience. “Digital” is assumed at this point. So, net-net: you must have a digital presence, and your digital presence is delivered via your application(s).

It’s just like having a cool website and doing some ecommerce, right?


Like many of us know, consumer-facing websites are no longer just landing pages with ecommerce capability. Mobile apps are no longer just for handheld content delivery. They are multifaceted applications that are made up of hundreds, if not thousands, of elements that range from code in a browser to ephemeral services to open source plugins, and even mainframe batch jobs. As a result, each of these elements requires the agility to match the fickle, ever-changing landscape of consumer demand and market trends. Fortunately, four key tenets of this tectonic platform shift have emerged to make this path to agility a possibility.

  • Cloud: The on-demand, utility-like paradigm brought about by the public cloud.

  • Open Source: The access, cost-effectiveness, and community of open source software.

  • SaaS: Best-of-breed technology options for when “buy” makes more sense than “build.”

  • Automation: The relentless pursuit to reduce time and risk as we adopt the above tenets and replace repeated human processes with workflow.

So, it began: the pursuit of microservices over monolithic code bases, of observability over reactive monitoring, of dozens of releases a day over one mega release per year. The tenets of the largest platform shift in technology history and these innovative pursuits ushered in new methods to design, build, and launch applications—a digital assembly line, if you will.

The Digital Assembly Line Requires A New Way of Managing Data

I’ll admit it, the general buzzwords associated with digital transformation tend not to be overtly data-centric. When we talk about application modernization or digital transformation, we tend to focus on concepts like microservices, ephemeral, functions, SaaS vs. PaaS, cloud native, container-based, etc. And yes, modern technologists are employing these methods as part of a digital assembly line producing a new era of agile applications.

So, what does this have to do with data? Wellat almost every point in the digital assembly line, there is a massive anchor in the shape of data that slows everything down. The modern digital assembly line is as much a lifecycle, and in some cases a reincarnation, of applications in an effort to instill agility. It consists of all the steps to build, maintain, and evolve what is now core to the digital experience today:

  • Development

  • Testing

  • Fixing

  • Patching

  • Re-Factoring

  • Re-Platforming

  • Error-Handling

  • Restoration

  • Forensics

  • Compliance

As technologists, we’ve witnessed the advancements in bringing applications to market—the rise of DevOps, the availability of IaaS/PaaS, the accessibility of open source, and so on. These concepts gave birth to a modern set of flexible tools and integrations that gradually simplified the layers of complexity plaguing the realization of a digital assembly line.

Well, all except one: data.

Data continues to be a merciless choke point at every step of the digital assembly line. It slows innovation, degrades quality, stifles performance, threatens resiliency, and prevents forensics. When you look at compliance regulations and the over-rotation towards data protection vs. data agility, it’s no wonder why data is often a bottleneck or roadblock to progress.

Making Data Programmable

Having built my career in Technical Services and Customer Success for some very innovative technology product companies, problem solving has always been part of our DNA. What a product could not do out-of-the-box fell to Professional Services. What a customer could not adopt without guidance fell to Customer Success. Well, as an industry, “data gravity” has been a persistent drag on innovation across all organizations - and the power at our fingertips to solve this problem is making data programmable.  This is an innovation - actually, it's a borderline paradigm shift - and it's a necessity.

We’re now in the age of programmable data infrastructure—where innovation demands that data be ephemeral and immutable. More specifically, it means harnessing the ability to microtize, migrate, and distribute compliant data across any platform anywhere via an open API framework. In doing so, we reduce and eliminate the "data gravity" plaguing cloud modernization, CI/CD optimization, and AI/ML realization. Much like the prolific opportunity to invent and reinvent today, this programmability of data is nothing short of astonishing.

Systematic data agility will then give rise to game-changing concepts like, for example, Production Operations, or “ProdOps,” where traditionally isolated workflows within the application, infrastructure, and security worlds combine with data to create a 1+1=10 effect.  This one might even go to 11. As an industry, we’re on the precipice of launching new waves of automation value, and this unleashing of data will be one of the linchpins that define success.

It’s time to invent and reinvent...again...and I couldn’t be more excited!


You can also read the original piece published on LinkedIn.