Capping IT Off

Capping IT Off

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

Alternate title: 

Thinking like a start up, lessons for the big IT department

Category : Digital Strategies
Can large corporations take a big leap forward from their current IT delivery model
 
For large corporations the benchmark for revolutionary rather than evolutionary change is to think and act as a start up would, leveraging state of the art services and tools. In this context I can see three strands of revolutionary IT change:
 
  1. Keep it simple rather than let complexity grow
  2. Automate rather than rely on labour arbitrage
  3. Business change rather than systems change
 
Keep it simple rather than let complexity grow
 
New service providers are focussed on having platforms that allow them to test new ideas and then discard the ones that do not work. In this way services can be continuously improved whilst ensuring the user experience remains clean and simple. Amazon is a leader in this area, driving continuous improvements through testing of ideas. For this to work platforms need to take functions out as much as they put new ideas in. This compares with a traditional approach to application management where changes accumulate over time so that complexity and consequently management and change cost also grows. A simple approach to start the journey is to put a cap on function points in operation so that any new services needs to be balanced by removal of existing services.
 
Automate rather then rely on labour arbitrage
 
Off-shoring of services can drive an environment of ‘your mess for less’ and remove the need to focus on automation. The DEVOPS war cry has stemmed for a lack of investment in corporations in the right automation of processes, systems, testing and tools. Start-ups operate with a different mantra where everything is automated from day one, meaning that there is complete trust in the process, releases can be done at will and an order of magnitude less people are required to deliver the services than in a traditional environment. This can require significant investment to engineer into an existing estate and takes time to build the right levels of trust in the process. However, once investment has been made the operations costs can be progressively removed.
 
Business change rather than systems change
 
Traditional IT systems require a process of requirements analysis and then implementation by the IT department. New platforms focus on moving more capability to be managed directly with the business avoiding the need for IT intervention. This is now progressing to the point where scripting tools allow the business users to directly create their own applications which when coupled with end to end automation means that they can also push it into production. For the start up today releases are many times a day rather than every few weeks and they are pushed directly by the business users through automated processes. The rapid development of SaaS applications which are sold directly to the business is pushing this agenda and creating an alternative IT delivery model.
 
This ability to use business level programming for creation of solutions is even coming now to the Internet of things and hardware devices through companies such as Sam Labs (https://samlabs.me/) who provide a ‘Lego’ approach to building solutions.
 
So how to progress
 
Our research with MIT (https://www.uk.capgemini.com/news/new-research-from-capgemini-consulting-and-mit-sloan-management-review-reveals-why ) on what are the key aspects of technology leadership in a Digital World illustrates that to achieve this step change in performance the IT organisation of the future needs to address 3 areas:
  • Building integrated business – technology thinking and relationships
  • Building new skills
  • Having the right digital technology platform which embraces the automation of processes.

About the author

Cliff Evans

Leave a comment

Your email address will not be published. Required fields are marked *.