Why Boring is not bad for Enterprises, Always

Background :

I joined my first job on 14th June, 2012, in Samsung Research, India as a R&D Engineer, as part of their Android Homescreen division. I had just finished by Bachelor’s in Computer Science, and I was eagerly waiting to test out all my theories, and concepts on a practical system, usher in new innovations, work with cutting edge technologies, and so on.

However, pretty soon I realized that the company did not work in that way. Yes, there were pockets of innovation, and experimentation in some isolated corners, but the core systems were handled more precariously, and many of my so-called ‘radical suggestions, code changes, approaches, etc‘ were not entertained with the same level of rigor, and enthusiasm that I had expected.

I spoke to my boss at that time, who had been in IT for close to 15+ years by then, and with whom I enjoyed a very cordial, and straightforward relation, on why this disdain with implementing some of the newer patterns, technologies, or approaches that I was recommending.

I expected him to say, the standard response that we, as Indians have been used to listening from our national cricket team coaches ‘If it ain ’t broke, don’t fix it’. Rather he said, ‘Boring is not bad for Enterprises, Always’ ( or something along similar lines ). And this one statement has stuck with me for the remainder of my IT career, and today I will delve a little bit into my perspective on the same.

?Are Enterprises Anti-Innovation ?

To a na?ve eye, the above statement from my erstwhile boss might seem like a declaration against innovation. But it is far from that. Yes, it is true though that enterprises might have to employ a more nuanced approach to radical technology changes, but only because they have more to lose.

Most likely, you will see that in the above curve, an enterprise will be more than happy to position themselves firmly in the ‘Early Majority’ section, especially for most of the enterprise products that are already being leveraged by clients, in production.

There will be isolated Innovation teams within big enterprises, who have a more radical approach, and will be in either the ‘Innovators’, or ‘Early Adopters’ section; however, they will most likely be working on POC(s), thought leadership, guidance documents, and industrialization of emerging offerings before production grade consumption.

Enterprises love stability, because their consumers love stability, but make no mistake, stability does not, and need not come at the expense of innovation.

And the Technology leadership of the company ( the C-suite ) is tasked with finding that right balance between both stability, and innovation, and that is why they get the big bucks.

Boring means money

Digital and Emerging technologies ( DET ) are cool, they attract new employees, and so on. One may get more press when implementing a simple Blockchain project, rather than a complicated Spring Boot distributed architecture; but at the same time, many of the major accounts that power an enterprise are on so-called “legacy technologies”, and will most likely continue to be the same, unless there is a real need to change.

In PwC, when I used to work as a .NET Architect, on a 5M USD+ project for a UK-based Insurance company, my team used to sit beside the DET team, who for the entire year, only performed POC(s) on Voice-Operated Chatbots, etc. We used to joke with their guys, that our projects are funding their Innovation.

It might not be as simple as that, but yes, cutting edge technologies might not bring in big deals initially, but at the same time, that does not imply, that team should be scrapped; the frameworks, and guidelines that they are building will most likely be the foundation of the industrialized solutions that clients will be able to trust, in a few years.

Fit for need

Amazon CTO Werner Vogels once said, “Building evolvable software systems is a strategy, not a religion. And revisiting your architectures with an open mind is a must”

Software systems do not exist in isolation, so where a monolith pattern might be more than enough today; there could be reasonable demands for a transition to microservice based architecture, when the number of developers in the team quadruple, as an example.

However, changes to systems, just because the technologies currently being leveraged are boring and not in the Thoughtworks Technology Radar ( https://www.thoughtworks.com/radar ), or because there is an over-energetic engineer ( Hint : Me, in 2012 ) who wants to try out the latest Metaverse technology on production, does not make any sense, either.

I like this cheeky quote from Ampt CEO Jeremy Daly, where he says

“DON’T build monoliths... except when you should.

?DON’T build microservices... except when you should.?

?DON’T use containers... except when you should.

?DO use serverless... except when you shouldn’t.”

When the right technology is leveraged for the right reasons, it is as sweet a feeling, as seeing Sachin Tendulkar hit a six off a Shoaib Akhtar ball; but here the keyword is ‘right reasons’. I don’t see a world were, emergence of new technologies can be the only criteria for refactoring or reorganizing a system.

Innovation, for the sake of it

Amazon Prime Video team made a lot of noise, when they published an article about how they reverted back to Monoliths from Microservices : https://www.primevideotech.com/video-streaming/scaling-up-the-prime-video-audio-video-monitoring-service-and-reducing-costs-by-90

“Moving our service to a monolith reduced our infrastructure cost by over 90%. It also increased our scaling capabilities.”, Marcin Kolny, Amazon

I saw multiple posts on Twitter, and LinkedIn where folks were debating on that age-old discussion on Monoliths vs Microservices, but I actually like the approach that the Amazon Prime team took. They tried an approach, and when it did not work, they tried something else, which made more sense, and they stuck with it. In this case, the approach which worked for them is “more traditional, or legacy”, whereas the approach which did not work was “more cutting-edge, or modern”.

But that is exactly the message that I want to spread via this article, Innovation just for the sake of it, is not the correct path, always.

Thoughtworks Technology Radar

Thoughtworks publishes a tech radar, where their analysts try to provide opinions on which technologies in the endless sea of ever-emerging innovation is in which state : Adopt OR Trial OR Assess OR Hold.

The fact that a company like Thoughtworks manages, maintains, and regulates this quadrant implies that not every single technology that has been released or available today, is ready for production consumption, at scale.

It is the responsibility of the Innovation Hub within the enterprise to work with the Technology Leadership and regulate this space and provide the foundation for the rest of the company to build applications on.

The Human Angle

As humans, we are naturally attracted to the shiny new things that are available in the ecosystem. And to be fair, the Innovation & DET teams appear to be the cool kids on the block, working with new stuff, presenting at conferences, getting recognition, etc; whereas the engineer who works on a Batch Job that reconciles 30M USD worth of transactions every day, feels underappreciated.

This is where the leadership needs to step in, preferably before any animosity or frustration deepens within individuals. Employees should be provided some sort of a pathway to transition to what is more aligned with their aspirations; and appreciations ( including monetary compensation ) should consider the full context of the enterprise setting, than just focus on the hype around new technologies.

Summary

In summary, I leave you with this quick excerpt from ChatGPT’s answer to the prompt : “Why boring is not bad for Enterprises, Always”

"Boring" is not inherently bad for enterprises; it can provide stability, reliability, and cost-effectiveness. However, enterprises must also be willing to embrace innovation selectively and strategically to stay competitive and meet the evolving needs of their customers and markets.

I love it when AI is aligned with my thoughts, and rest assured, this is the only part I borrowed from ChatGPT ( but no way to verify though, just kidding ).