Can data analytics inform new business model development?

Information technology has enabled the upheaval of longstanding business models, whether in the globalization of product supply chains, the outsourcing of manufacturing, delivery and support processes or the use of efficient, just-in-time manufacturing. The list goes on.

While the changes have many benefits, they have also served to make the product development process more complicated, with one notable exception: the supply chain simplification from disintermediation.

Disintermediation, namely the elimination of middlemen, took off with online commerce, but the notion of more directly connecting customers and suppliers has started to seep into almost every business sector. While disintermediation has historically involved streamlining a linear supply chain, the next stage in business evolution entails creating bidirectional connections between various stakeholders in the product development, delivery and purchase process. That’s the thesis of Hortonworks’ CEO Rob Bearden who characterizes the evolution of linear supply chains that follow a set of procedural processes into a mesh of connected communities comprised of customers, suppliers, producers, manufacturers and service providers as

…the biggest business model transformation since the industrial revolution.

Hyperbolic? Perhaps, but Bearden used his keynote at the Dataworks summit and a separate presentation for analysts to expound on his proposition and explain how the explosion of business data and the increasingly sophisticated means of analyzing it are behind this shift.

From data silos and batch processing to data lakes and real-time analysis

The contributors to data-fueled business processes and decision making are by now familiar. The falling cost of storage, the addition of intelligence and connectivity to all manner of industrial and consumer devices, pervasive mobile networks and device usage and dramatic improvements in data science hardware performance and software capabilities.

These are the technical component elements that make the creation of two-sided markets of the kind Beardon describes and which the likes of Uber and Facebook have utilized in the development of their business models.

The ramifications are equally familiar from eerily tailored online ads curated from social media activity, dynamic pricing for airline tickets and hotel rooms or customized grocery discounts based on past purchases.

Each of these stems from the ability for organizations to aggregate data from disparate sources and use sophisticated machine learning techniques to not only summarize historical events but predict trends and identify previously obscure correlations.

Big data systems like Hadoop are the foundation for such data unification. However Bearden identifies four other catalysts for what he calls data-driven business models (in order of market size and significance):

  • Cloud services
  • AI, particularly ML software
  • Streaming, real-time data
  • IoT devices (which are often the source of data streams)

In a blog post summarizing his keynote theme, Bearden describes data management practices as stemming from the increasing diversity of enterprise data types and sources.

Until now, enterprise data was largely structured and not particularly diverse. This reality spawned what we know now as traditional IT best practices: buy standardized packaged solutions, or use them to replace custom solutions and implement an EDW based on user requirements.

As storage technology and data science progressed, business innovators discovered the value of eliminating silos of information, aggregating data into massive repositories, i.e. data lakes and turning a new generation of ML algorithms loose to see what might be revealed. Hortonworks’ CTO Scott Gnau calls these “clouds of data.

Bearden characterizes the ramifications this way (emphasis added),

The implications of being able to connect data and software in this way [ aggregating all data, at-rest in the data center and streaming from devices], are huge value creation and transformation for businesses. … Once you remove the silos, become more data and less app-centric, you can start to do analytics in real-time as data is streaming. So rather than running a report to see what happened and then having 17 meetings internally to determine what action to take after the fact, and it’s too late, you can move the time of decision right to when your people are interacting with the customer. In other words, to the point when the decision needs to be made. In some cases even being pre-emptive before an event happens.

From linear supply chains to connected business communities

The endpoint is what Bearden calls connected communities in which enterprises, their customers, suppliers and logistics providers (supply chain) become sources of information for each other. Such direct communication and data sharing, which I contend is a form of disintermediation, replaces a linear product-to-consumer process with a bi-directional mesh of business stakeholders. Bearden contends that these new business models provide real-time visibility into product performance and customer satisfaction that enables rapid response to product plans, customer problems, and feature requests.

Writing in his blog, Bearden describes the business consequences of such dynamic, interconnected business models (emphasis added).

The impact of these changes is profound across every industry. In retail, for example, the ability to analyze customer buying behavior down to the level of the in-store beacons or smart hangers and being able to map this in real-time to social signals and online visual search is providing a 360 degree view of the customer that fundamentally changes the concept of a loyalty program. … Tie this to the inventory system, store operations, and real-time supply chain optimization from POS systems, and retail businesses can start to transform how they do business. Look at the results some companies are already achieving: doubling promotion revenue lifts, cutting fraudulent transactions and storage spend in half.

Gnau believes that as enterprises, their suppliers and customers realize the value of the data they provide, seeing it as an asset, not an extraneous aftereffect, each will seek ways to monetize it. While he doesn’t offer any mechanisms for doing so, one can see the implications of customer monetization in the increasing sophistication and customization of retail rewards programs.

Retail, along with logistics, are two areas in which the consequences of massive data gathering and analysis are easiest to see. Amazon is the quintessential example of a data-driven company, but it has also caused other retailers to up their game by exploiting their many data sources and customer interactions. In a recent interview, Anuj Dhanda, the CIO of Albertson’s, one of the top-three grocery retailers in the U.S. detailed how it is using data and IT services to streamline the supply chain and customize product selection.

Our supply chain, from the manufacturer to the customer, is getting more modernized and customer-centric all the time, and this is where data is really going to make a difference.

Dhanda added that the company plans to use its ample stockpiles of customer data along with IT automation software to create a touchless experience for gas station customers akin to the Amazon Go, grab-and-go convenience stores. Amazon Go uses cameras to identify customer vehicles, link to customer rewards account and credit cards to provide any earned discounts and complete the sales transaction without running through a point-of-sales system.

As someone who shares a hometown with Albertson’s and has been a customer for decades, I can attest that its technology-driven marketing, from its mobile app to customized, in-store discounts, has significantly improved over the past couple years.

Bearden also pointed to Trimble, a supplier of navigation systems and software that does significant business with transportation and logistics companies, as another company that has used the real-time analysis of data to significantly improve its offerings.

According to Trimble’s CTO, big data and analytics allow it to provide real-time vehicle location and shipment data, financial planning based on shipping revenue and fuel cost and faster payments for more than 4 million trucks a day. These capabilities have cut the manual work required to manage shipments in half and contributed to double-digit revenue gains for Trimble that have provided the capital to invest in new BI and data analysis products.

My take

Whether you call it digital transformation, business model transformation or, as Amazon founder and CEO Jeff Bezos puts it, “unrelenting customer obsession, ingenuity, and commitment to operational excellence” in the face of “ever-rising customer expectations,” there is no doubt that various forms of IT have catalyzed and powered new efficient, adaptable, responsive ways of doing business.

The access to and ability to intelligently process massive quantities of both historical and real-time data undoubtedly contribute to these changes. However, as I wrote last week about Hortonworks’ corporate strategy, technology companies have an inherent tunnel vision when it comes to their area of focus and expertise, seeing every salubrious effect as stemming from their noble cause. As the saying goes, “when all you have is a hammer, every problem looks like a nail.”

I contend that network technology, both terrestrially via the Internet and vast build outs of fiber capacity (some still dating from the premillennial dot-com boom) and wirelessly via ubiquitous high-speed cellular networks married to phones that have evolved into multi-functional devices, is as significant.

A defining attribute of the new era of business intelligence is the ability to collect, aggregate and analyze, both centrally and locally, based on previously developed models, enormous quantities of data. Without networks, such data would remain in silos or lost as noise from the growing cacophony of devices spewing it out. Another type of network, deep learning neural networks, has become a critical means of unlocking previously obscure insights from seas of data.

Thus, I agree with Bearden in part that data enables a new era of business models. However I believe it’s one side of a foundation that’s also composed of:

  • Global, software-defined terrestrial and mobile networks
  • Massive, efficient cloud infrastructure and IT services
  • Adaptive, self-optimizing AI and data science algorithms using ML and deep learning

Image credit – via the author Disclosure – Hortonworks covered the majority of the author’s T&E for attendance at the customer event.


Article by channel:

Read more articles tagged: