Bimodal IT, How to Balance Stability and Agility?

Cover Bimodal IT

Bimodal IT, How to Balance Stability and Agility?

Share on twitter
Share on linkedin
Share on pinterest
Share on reddit
Share on pocket
Share on email

Over the past few years, the revolution in the digital world has led to and forced numerous reorganizations, including the arrival of the “bimodal”, which poses the following problem: “how to bring together two organizational modes at the opposite ends of the spectrum”. Answer: coordinate and reconcile two different but not incompatible visions. To quote Dave Aron, Vice President of Gartner, “Delaying the adoption of bimodal IT is the worst thing a CIO can do”.

Bimodal IT: Two Modes of Data Management

Let’s start with a definition of what “bimodal” is. Bimodal is simply the existence of two distinct but possibly coexisting data management models.

According to Gartner, bimodal IT is “a practice of managing two separate and coherent modes of IT delivery, one focused on stability and the other on agility. Mode 1 is traditional and sequential, focusing on security and reliability. Mode 2 is exploratory and non-linear, focusing on agility and speed.

The real challenge lies here: how to allow the optimum cohesion between these two types of approaches, which are so different but yet so complementary. What is the objective? To reconcile these two working rhythms.

However, it is clear that these two models are compatible and that a collective work would allow a complete and successful digital transformation. Both modes are necessary for smooth operation of the company and the creation of added value. Succeeding in its digital transformation is fundamental to the survival and development of the company in the digital world.

According to Gartner, mode 1 consists of the traditional IT, the IT department, and is based on a very linear approach in which foresight, accuracy and reliability prevail. Mode 1 likes accuracy and security. Understanding and planning are therefore essential and are the basic working method of mode 1 . Indeed, the entire stability and industrialization of the company relies on the IT department.

Mode 2 represents “new generation” Cloud-based computing. This type of organization is younger and more dynamic, “in Paas mode”, with an acute sense of innovation, agility and speed. It meets the needs of the automation culture. The level of uncertainty of this model is quite high and, unlike mode 1, is of very little interest. Mode 2 likes to explore, try and experiment!

Basically, mode 1 is based on reflection and preparation: think before you act, check that everything is well programmed with the objective to produce. Mode 2 is based on POC (Proof of Concept), which is a method for carrying out a concrete and preliminary experimental project in order to prove its feasibility. Mode 2 prefers to act in the short and medium term: by observing the results it will be able to renew the experiment, improving it and adapting it to the situation. Acting quickly, “innovating to move forward” is its leitmotiv. It is the advent of the “quick fail” that allows to respond quickly to the pressing expectations of the businesses.

The major difference lies particularly in the vision of each individual. Indeed, mode 1 only sees in the long term. Building a Big Data project is a very long process and can be spread over several months or even several years. But mode 2, which is more dynamic, wants to go fast and act with agility. It is true that speed of execution represents a real competitive advantage, especially in a world that is constantly evolving and demanding ever greater speed.

I will therefore conclude this first part by defining mode 2 as “wild” and “innovative”, as opposed to a more traditional mode 1, advocating reliability, expertise and safety.

Yes, as you will have understood, these two management models are very disparate, but one needs the other to move forward and vice versa. The challenge is to find the right balance between the two. What do these two modes have in common? The customer’s goal.

Agility as a Key Principle for Modernizing IT Infrastructure

Agility, or “do it as a startup”, is a basic principle to innovate and act quickly. It is therefore necessary to move away from traditional sourcing methods in order to boost the company’s agility in a sustainable way. Unfortunately, so-called traditional methods, especially for large accounts, tend to limit the actions of improvement and innovation, whereas adaptive sourcing methods allow more agility in the company.

The answer to this challenge 2.1?

A Big Data platform that would perfectly meet the needs of each of the two modes and that would act as a bridge between mode 1 and mode 2.

What do they need? To analyze data and put it into production quickly to maximize the customer experience. The challenge? Combining agility and control.

To successfully break into the digital world, filled with possibilities and technologies each more futuristic than the others, it is necessary to achieve its digital transformation, a crucial but far from simple step for a company. Indeed, the opportunities but also the dangers that the new digital world brings push companies to favor innovative, fast, flexible and collaborative solutions, in which the different IT worlds can meet and move forward together with the same tools and objectives.

The bimodal company, whose objective is to accelerate its digital transformation, must first of all adopt an approach adapted to the two organizational modes that will combine their skills and adapt their management methods to each other. The aim is to find a way to reconcile the two models and enrich the value chain.

Digital transformation is now at the top of the agenda for companies who are now turning to “digital” and Big Data to maximize the value contained in their data. Studying data means having access to billions of previously unoptimized information.

In fact, CIOs in particular are the ones who want to accelerate their transition to digital. But to achieve this, close collaboration with traditional IT is essential. So the solution would be to create a common structure for the two modes, which would encourage the development of their work tools for greater cohesion, and which would therefore provide CIOs and Data Scientists with the same technologies, in other words a bridge between creation and industrialization.

Indeed, everyone will agree that it is impossible, or at least very difficult, to push into production projects that do not use the same technologies from the beginning to the end of the production chain.

But how can the product be put into production and industrialized quickly and in total coherence with the producer? How can data extraction and analysis tools be pooled to create applications that adapt to the business and the end user?

A platform to experiment freely, test and put into production in a minimum time many Data Science treatments in a standardized way. Coordinate product development and industrialization.

A Big Data platform that would perfectly meet the challenges of both modes by linking agility and security.

The key word is “governance”. The governance of your data allows you to classify, gather, name and qualify them in the same way, to standardize them and make your data lake more homogeneous and readable. When access to information is simplified, everyone then has the same tools, which can be understood and used by everyone.

Moving from IT to digital is not simple and requires the acceptance of many disruptions in entrepreneurial organization and management methods. The CIO’s approach is flexible, mobile and contextualized and must complement the more structured, precise and controlled vision of traditional IT. We must therefore succeed in balancing agility and, at the same time, reassure traditional IT.