by Roy Atkinson
Date Published September 7, 2016 - Last Updated December 15, 2016

In July 2014, Gartner published Bimodal IT: How to be Digitally Agile Without Making a Mess. In September 2015—barely more than a year later—a Forbes headline exclaimed Bimodal IT: Gartner’s Recipe for Disaster. In January 2016, CIO Insight published an article titled Saying Goodbye to Bimodal IT. That’s a pretty steep spike in Gartner’s own hype cycle. What happened?

The idea of bimodal or two speed IT seems very straightforward: Agility and velocity are attributes desired—if not required—in today’s digitally driven businesses. The current popularity of DevOps is witness to that. But companies in highly regulated verticals such as finance, insurance, and healthcare often have huge technical debt (i.e., the burden of legacy critical systems that cannot be taken offline without enormous effort and that present enormous risk in case of failure). Because of this divide, the bimodal model seems very logical: Be slow and careful with the critical and often legacy infrastructure, but be fast in delivering user and customer facing applications and services.

Want to know more about DevOps? Attend the DevOps FUSION preconference and the DevOps and Agile track at FUSION 16.

DevOps icon Jez Humble wrote, in The Flaw at the Heart of Bimodal IT, posted in April 2016:

There are three serious problems with the Bimodal model which, when taken together, mean that leaders that fail to move beyond Gartner’s advice will end up falling further and further behind the competition. They will continue to invest ever more money to maintain systems that will become increasingly complex and fragile over time, while failing to gain the expected return on investment from adopting agile methods.

Humble goes on to talk about the three problems, which he lists as:

  • Moving from a one-size-fits-all to a two-sizes-fit-all model
  • Forgetting that the rapidly developed systems usually tie back to the legacy systems of record
  • Thinking that responsiveness and reliability are mutually exclusive

As many proponents of DevOps will tell you, the goal is not just to be fast; the goal is to build secure and highly stable systems by rapid iteration and continuous delivery. Why shouldn’t these principles be applied to the systems of record as well?

Bimodal IT was an attempt to get people moving into transformations and sought a way to do that without scaring more traditional CIOs and directors into next Tuesday. Re-engineering the basic applications your company is built upon is a high-risk proposition. Millions and millions of dollars can be lost and reputations with them.

There’s a story, probably apocryphal, that says a COO who was concerned about spending a lot of money to train a group of engineers once said to the CEO, “What if we train them and they leave?” The CEO is said to have replied, “What if we don’t train them and they stay?”

In similar fashion, we can ask, “What if we try to replace it and it fails?” Maybe it’s better to think, “It’s going to fail. We better fix it carefully—and fast.”


Roy AtkinsonRoy Atkinson is HDI's senior writer/analyst, acting as in-house subject matter expert and chief writer for SupportWorld articles and white papers. In addition to being a member of the HDI International Certification Standards Committee and the HDI Desktop Support Advisory Board, Roy is a popular speaker at HDI conferences and is well known to HDI local chapter audiences. His background is in both service desk and desktop support as well as small-business consulting. Roy is highly rated on social media, especially on the topics of IT service management and customer service. He is a cohost of the very popular #custserv (customer service) chat on Twitter, which celebrated its fifth anniversary on December 9, 2014. He holds a master’s certificate in advanced management strategy from Tulane University’s Freeman School of Business, and he is a certified HDI Support Center Manager. Follow him on Twitter @HDI_Analyst and @RoyAtkinson.


Tag(s): future of support, devops, service management, supportworld

Related:

More from Roy Atkinson


Comments: