Stop assuming that the answer is to simply install Agile, DevOps or Cloud

Publish date:

Organizations have to move with the time – Digital, Cloud and DevOps are trends that are shifting and changing how enterprises are organized and speed and value for money are key driver of change. Many look up to the “Cloud & DevOps Masters” and assume they can simply apply a “Cloud & DevOps solution” to […]

Organizations have to move with the time – Digital, Cloud and DevOps are trends that are shifting and changing how enterprises are organized and speed and value for money are key driver of change. Many look up to the “Cloud & DevOps Masters” and assume they can simply apply a “Cloud & DevOps solution” to achieve same/similar outcomes. But, there is a snag; most organization cannot just move from a traditional (some call it legacy) to a Cloud and DevOps-based setup—it is a transformation journey and not just a simple step installing “a solution.”

The majority of organizations I have worked with over the past 5 years are very large, global with 1000+ applications (packaged traditional, custom package and custom build, mix and full web-based as well as cloud native), who have outsourced some of their IT to a 3rd party and who’s retained IT organization is organised against application and infrastructure-related criteria.

Changes are typically packaged up and are being deployed in large, complex and difficult releases every month and the retained IT organization is being measured against annual cost saving targets, number and duration of service outages and maybe internal end user satisfaction. However there is no or little direct business interactions nor any real end to end business understanding on the IT side.

Propelling such an organisation into a setup that can deliver change in an almost instant, while constantly reducing outages will require a step-by-step plan that follows an IT maturity model related to three core areas: people, process and tools.

In detail the following characteristics are apparent for Level 1,2 and 5:

Level 1: Basic

  • People: separate strategy, design, development, testing and live operations teams. Complete lack of terms of references. No joint sessions, get-togethers. Teams focus on their own direct targets and objectives only. No joint or shared objectives and no overall reward system. People only feel accountable for their immediate area—no common or overarching ownership.
  • Process: separate and disconnected processes are place which are ah-hoc, reactive and chaotic. No common end-to-end process framework, no common signoff criteria or any joint solution design characteristics that support appropriate “–ilities” (availability, stability, flexibility)
  • Tools: no automation tools, majority of activities are manual, ad-hoc and unplanned. No integration between hardware provisioning, operating system installation/configuration and middleware/application related provisioning/installation. No sharing of joint configuration information with all information being stored and retained in different repositories.

Level 3: Co-ordinated

  • People: mainly silo’ed organization however lead architect/lead designer(s) increase their scope to also include operational aspects. Joint sessions are held to increase wider visibility—for instance key operational staffs are actively engaged in the design and build phase. Developers are also measured on operational characteristics.
  • Process: still mostly separate processes covering the entire solution lifecycle however there are some joint process points where development and operational aspects are jointly covered. Better understanding of the entire environment setup and characteristics.
  • Tools: most of the development environment setup is being created automatically. Only application-related components are manually installed.

Level 5: Top

  • People: one team, co-located and extensive collaboration and knowledge sharing.
  • Process: single overall process covering the entire solution lifecycle—from strategy, planning to design, build, test to run.
  • Tools: all environment setup are being created automatically from a single repository. This covers all aspects—servers, operating system, operating system near as well as all middleware and application related components. No manual processes in place.

For most organizations the step by step plan will result in a number of key changes, all geared towards speeding up and increasing quality:

As noted above typically organization exhibit a basic maturity—this typically means that they are follow a silo’ed and tower-based structure, with a mostly manual IT setup that does not provided shared capabilities and only delivers some limited aspects as a service typically.

To increase its maturity (with the main goal to increase speed and quality) from Level 1 to Level 3 an organization should focus on

  • reducing complexity by consolidating its application and infrastructure capabilities;
  • to start developing shared services and;
  • to move from waterfall to agile and establish a non-production platform approach;
  • reduce manual activities and establish more “as a Service” capabilities.

The step to Level 5 might be direct or in some cases indirect—some organizations might stay at Level 3. To move from 3 to 5 the following aspects are key:

  • establish both a platform capability for non and for production;
  • use a cloud first/only approach;
  • embrace Agile [3] and DevOps [3];
  • reorganise to align with business;

The above only related to the “what.” To understand the exact sequence as well as scope of each transformation step(s)—the “how” typically you will need to have:

  • established a comprehensive IT Strategy [1];
  • understand where you are today and what the target should be;
  • setup and execute a joint business and technology transformation program.

These are technology changes that affect the business the same as the IT organisation so key to success is joint up working that allows for both top- and bottom-up sequencing.


The solution for many organization that need to speed up whilst reducing outages can be Agile, DevOps, Cloud, automation, “as a Service,” etc. However, you cannot just deploy some Cloud or DevOps solutions and expect that the solution lifestyle and outage pattern will be at par with Amazon or Netflix, etc;

For one Cloud, Agile and DevOps are not solutions; they are an approach, an philosophy related to people, process, tools and technologies. Secondly, each organization is different and the pace and sequence of change will impact the speed of transformation. Assuming an organization is at Level 1 it cannot simply jump to 5—a transformation journey is needed. And lastly the only constant is change, so each transformation journey must itself adapt new approaches and tools.

Thanks for Reading. 

About the Author: Gunnar Menzel has been an IT professional for over 25 years and is VP and Chief Architect Officer for Capgemini’s Cloud Infrastructure Business. According to Richtopia Gunnar is one of the Top 50 most influential Information Technology Officers. His main focus is business-enabling technology transformation and innovation.

References: [1] [2] [3] Agile, DevOps & Lean Unlocked – The perfect combination to accelerate or a recipe for confusion?

Related Posts

accounting automation software

Envisioning the future of accounts receivable

Date icon October 26, 2021

Capgemini is working together with BlackLine to build a smart, and flexible, and frictionless...


Sustainable IT – attention needed

Date icon October 8, 2021

Advances in technology means environmental challenges are likely to intensify. Few...

Business Process Outsourcing

Simplifying HR operations through automated desk booking

Date icon September 29, 2021

Capgemini’s HotDesk tool gives our teams the confidence to return to the office safely and...