The 5 pillars of digital transformation

October 3, 2019 Sutha Kamal

The world has never moved as quickly as it does today, and it will never again move as slowly as today. In the past several decades, the average tenure of a company’s listing on the S&P 500 index dropped from 60 years to just 20 years, and it’s getting shorter. It’s this acceleration—as well as the disruption that comes along with it—that has centered the most important boardroom conversations around digital transformation.  

At its core, digital transformation is about software. There has never been a technology more powerful than software; it is a force-multiplier for the human mind. We can scarcely imagine designing, making, or marketing anything without software, and it seems obvious that their prowess in wielding software is what has enabled the digital natives to disrupt incumbents. 

To quote Steve Jobs:

"I think one of the things that really separates us from the high primates is that we’re tool builders. I read a study that measured the efficiency of locomotion for various species on the planet. The condor used the least energy to move a kilometer. And, humans came in with a rather unimpressive showing, about a third of the way down the list. It was not too proud a showing for the crown of creation. So, that didn’t look so good. But, then somebody at Scientific American had the insight to test the efficiency of locomotion for a man on a bicycle. And, a man on a bicycle, a human on a bicycle, blew the condor away, completely off the top of the charts.

 

"And that’s what a computer is to me. What a computer is to me is it’s the most remarkable tool that we’ve ever come up with, and it’s the equivalent of a bicycle for our minds.”

Advice abounds on how to transform, with themes like customer journey and experience, lean, personalization, the move to services, DevOps, digital twins, and more. These are all important, but you want to be certain that each dollar you spend on digital transformation is spent in the best possible manner. You need a plan for making the highest-leverage investments.

However, in all this technology talk, we often overlook that time and the creative potential of employees are the only truly scarce resources, and software is merely a force-multiplier for that creative potential. Money has never really been scarce. With a compelling vision and a credible plan, there’s always capital available. But vision and credible plans are rare, and they come from the creativity of your employees.

So the most important part of a digital transformation is the use of software to unleash the most creativity, across the most employees. Because we can’t buy time, the second biggest contribution is to accelerate the pace at which those employees can take their insights from “concept to cash”, or from idea to production. I believe there are five pillars of digital transformation that smart organizations can adopt in order to unleash this creativity and accelerate the pace of product rollouts. Each transformation dollar invested in one of these pillars is well spent.

1. Modern business strategy and processes 

The function of “strategy” needs to evolve in this complex and accelerating world. For most businesses, the days of 10-year plans are long over because it’s impossible to know what the world will look like that far out. 

As I write this, for example, the United Kingdom is on its way toward crashing unceremoniously out of the European Union; uncertainty about a previously unthinkable Sino-American trade war rattles markets bracing for a global recession; and Verizon recently sold Tumblr for $3 million only 6 years after Yahoo paid $1.1 billion for it. Some of our predictions about the future will certainly come to pass, but many others will not. 

In today’s climate, making business decisions that rely on distant projections is imprudent. The problem with having a grand strategy with a singular focus means you need to be right. Every. Single. Time. It’s like planning for retirement by spending your life savings to buy a single, and incredibly expensive, lottery ticket.

Once upon a time, having the “killer app” was enough; today, you need a portfolio of killer apps.

In the past, we surveyed the landscape, made predictions about customers and competitors, and revered focus above all: Our mantra was, “Have one plan and execute.” Today, optionality is the game. Twenty years ago, it was enough to have a “China strategy”; today you need five. Once upon a time, having the “killer app” was enough; today, you need a portfolio of killer apps.

For disrupters and transformers, the strategy is to create portfolios of options (hypotheses and experiments), then accelerate the speed of imagining, experimenting, learning from, and re-imagining these experiments. Steve Blank, in The Four Steps to the Epiphany, and later “Why The Lean Startup Changes Everything”, sums up the limitations of old strategy, and provides what is now the blueprint for disruptors and transformers alike.

2. Modern software tools and processes

In the same way, although software is nothing new, it’s now obvious that mainframes and waterfall development can’t compete with cloud infrastructure and agile development. It’s not that mainframes and waterfall never worked, but rather that they don’t work anymore. They are neither fast enough nor flexible enough to keep up with what the market requires in terms of functionality, or the creative solutions your employees develop to address those needs.

Adopting modern software development tools and processes—cloud-native tools and architecture, along with agile, DevOps, CI/CD, and automation processes— are fundamental to doing digital transformation right. Across the board, they are designed with the goal of increasing the velocity of code changes and making it simpler to manage increasingly complex applications. In a software-driven world, products improve continuously, not quarterly, and winning organizations embrace this mentality by empowering developers—as well as security, operations, and other IT personnel—to move quickly. 

 It’s not that mainframes and waterfall never worked, but rather that they don’t work anymore.

Disruptors like Amazon, Google, Airbnb, and their peers became household names and, in some cases, Fortune 100 companies, in less than 20 years. At their core, they are software companies. Their success relies on moving quickly to capture new opportunities and build products that continuously improve (and stay online), faster than an accelerating competition. Software engineering doesn’t merely support the business; it is the interface between your employees minds and your business. It is the business.

3. Collaboration 

The boundaries of the corporation are blurring. Being able to collaborate both inside and outside your organization—with external vendors and partners as seamlessly as with employees—is a competitive advantage. However, two minds are better than one only if collaboration is seamless. 

If your competitors are using WebEx and email attachments to collaborate, while you’re using Zoom and shared Google Docs, you’re moving faster and involving more creative minds. When your employees can focus on the things they’re best at, and leverage partners to do the things that they are best at—and they can do this faster and cheaper than your rivals can—then those rivals can’t compete.

The startup that is seeking to disrupt your business is leaner, more agile, and is able to work effectively with many more partners and vendors because they’re better at collaboration. They know that the edges of the business are blurry, and that they can go faster by working with external experts and partners, not doing everything themselves. Your IT organization should be doing the same if the company aims to stay competitive for the long term.

High-performing organizations know how to collaborate, and it’s why they accelerate their usage of open source software over time. 

Open source software is another great area for competitive advantage, and it is fundamentally about collaboration. Kubernetes has more than 6,500 individual contributors, and TensorFlow nearly 10,000. Open source software projects are not mere code, they are social networks, which should explain why Microsoft paid $7.5 billion to acquire Github: it’s the social network-of-networks, the collaboration platform for open source software. 

In their book Accelerate, authors Nicole Forsgren, Jez Humble, and Gene Kim find that high-performing software organizations are nearly twice as likely to be extensively using open source software as are their lower-performing counterparts. These high-performing organizations know how to collaborate, and it’s why they accelerate their usage of open source software over time. Humans are social and creative, and collaboration breeds collaboration.

4. The data-driven enterprise 

Creativity is the most scarce resource you have, and decision-making is the most valuable place to apply it. The highest-leverage thing a human can do is to make a decision. From creativity comes decisions, from decisions come actions, from actions come knowledge, learning, and again creativity and new decisions. 

We must become data-driven organizations because doing so helps get this flywheel spinning more quickly.  Becoming a data-driven enterprise isn’t a single outcome, but rather a commitment to a continuous evolution through four phases. And because data enables continuous learning, there is no end to it.

Collection

The first phase is the collection phase. This is where we build the awareness up and down the organization that data is important, and can be collected. This is when we begin to instrument the business, collecting information we never had before. For example, before you can optimize your production line, you need to understand how it actually operates. And to do that you need to know what the people and machines are doing at any given time.

Aggregation

The second phase is the aggregation of data, so it’s usable. You might have the most sophisticated robots, or the most data flowing back from those robots, but if each of those data streams is in a separate silo, or isn’t easily available to your employees, it’s not doing you any good. Data lakes and stream-processing platforms like Kafka are examples of technologies that make data available and useful to the business. 

Putting data in the hands of developers is also profoundly powerful: Queries aren’t limited to the C-suite, and IT and the business can begin to digitize, quantify, and understand previously opaque business processes. 

Querying

However, while data lakes and Kafka are great for developers, extracting business value from data means giving a broader set of employees access to this data. So the third phase is the querying phase. Business intelligence and analytics tools like Tableau and Looker are good examples of relevant technologies for this phase.

Evolving to this third phase is no small accomplishment, and it means that a large fraction of your employees can ask questions of data to make faster, higher-quality, data-driven decisions. Consider this scenario: 

Your business makes Internet-connected manufacturing robots that send back telemetry about their operations and health. Your curious customer service rep has gone through the data and found the signals that predict future customer unhappiness. On further inspection, she determines that these signals actually predict machine failure and that, for some use cases, the solution is just to change the coolant in the machines twice as often. She creates a dashboard that shows her when a customer might run into a problem and, when it flags one, she goes to one website to schedule a service-call, and then to another to order the coolant to be sent to the customer before the machine fails. The customer is happy, the machine didn’t fail, and both companies have saved all the time and cost of fixing a problem after the fact.

This is great, but having exercised her creativity, your customer service rep is now stuck doing a bunch of manual, repetitive steps each time a customer is at risk, and that’s not the best use of her time. 

The programmable business

Now we evolve to the fourth phase: making the business and its processes programmable (API-ifying them, more specifically). Instead of a dashboard, with APIs the customer service rep can define “if this, then that” actions that programmatically schedule the service call and ship the coolant without her intervention. She can even set another rule to send a message to the account executive each time this happens, and even send an annual report just before contract-renegotiation time showing the sales rep how many times the customer was “magically” saved from downtime without ever knowing it. 

At the third stage of evolution, your employees’ creativity has been unlocked because they can better understand the business and they can make better decisions, faster. At the fourth stage, you’ve further freed their time (allowing more room for creativity), and replaced repetitive value-destroying processes with software and automation. At this stage, human decision-making (for some decisions) can run at software speed and scale. Digital disruptors are born in the third and fourth phases of this evolution.

5. Relentless democratization of technology

If software is a bicycle for the mind, it’s important to understand that there are many different kinds of bicycles for many kinds of riders. Some developers are riding powerful dirtbikes on tricky terrain, at the razor’s edge of machine learning and augmented reality. Other developers are riding all-terrain bikes—standard issue, reliable, and safe. They’re using a platform to let them focus above the value line. Employees who work outside of IT might benefit from training wheels to help them get started riding.

By 2017, it became clear that machine learning was a breakthrough technology, delivering previously inconceivable improvements in solving difficult problems like image and speech recognition, self-driving cars, risk modeling, and medical diagnostics. From tech executives to world leaders, people realized that mastering this technology would change the nature of competition between corporations and nations. But even at places like Facebook and Google, which already housed the most in-house machine learning expertise, the overall percentage of employees who could meaningfully use these technologies was still very low. As a result, these disruptors rapidly built and scaled “bootcamps” to put this powerful new technology in the hands of more of their developers. If a technology is a force-multiplier, putting it in the hands of the most people is most valuable to the business.

What about going beyond the developers, and teaching even more employees to use software to their advantage? Airbnb, a digital disruptor with a rich data infrastructure (by and for developers), launched Data University to empower more of the business to use these tools and ask questions of their data. These courses were immensely popular, with 6,000 class registrations in a company of about 4,000 employees. If Airbnb’s business staff can ask complex questions about the business and make rapid data-driven decisions, can their competitors afford to keep IT and data in a silo for the privileged few? 

If a technology is a force-multiplier, putting it in the hands of the most people is most valuable to the business.

These types of bootcamps, data literacy programs, and other learn-to-code efforts, combined with powerful, readily available software and cloud services create fertile ground for unleashing employees’ creativity. Compare that to the first wave of business intelligence products (Cognos, MicroStrategy, etc.), which were massive, expensive, complex, and brittle enterprise applications that could answer only a few questions, and only for C-level executives. 

Think about what different technologies can you bring to your organization to unlock more creativity. Perhaps it’s a training program for developers, an analytics platform for the business, or a Functions-as-a-Service (FaaS) platform so your expert non-developers (like data scientists) can easily deploy their AI models to help the business. Here are some examples of investments you can make today to put the most powerful tools in the hands of your entire organization.

Conclusion

Software is eating the world, and digital natives and successful transformers put software at the core of their business, and optionality at the core of strategy. They collaborate more, across and beyond the walls of the organization. They collect and use data to make better decisions, faster. They are continuously bringing powerful technology to a broad group of users, not keeping it in the hands of the few. 

To survive, incumbent businesses must at least improve their software development capabilities. To thrive, they must reorient around a simple truth: Time and employees’ creativity are the only truly scarce resources. Businesses must work to unlock the most creativity from all employees, and make faster, better data-driven decisions. 

I hope the Five Pillars serve you as a guide, helping you make the highest-value investments along your transformation journey.

Further reading

Coase’s theory of the firm [The Economist]: This helps to explain why reducing the costs of collaboration between individuals changes the nature of the boundaries of the organization. It’s academic, no doubt, but is worth understanding, because it’s one of the defining changes of the modern economy. 

Why the lean startup changes everything [Harvard Business Review]

The Lean Startup [book]

Why software is eating the world [Andreessen Horowitz blog]

Phoenix Project: A novel about IT, DevOps and helping your business win [book] 

Accelerate: The Science of Lean Software and DevOps: Building and Scaling High Performing Technology Organizations [book]

The CIO’s guide to CI/CD [Intersect]

The CIO’s guide to Kubernetes [Intersect]

Microservices are not the destination [Intersect]

An introduction to event-driven architecture and Apache Kafka[Intersect]

Software is blurring the line between tech and retail, new and old [Intersect]

No Previous Articles

Next
The next generation of SQL is about simplicity and global scale
The next generation of SQL is about simplicity and global scale

Cockroach Labs CTO Peter Mattis explains the new breed of distributed SQL databases.