BiModal IT and Other Snake Oil – Part 2.

So, in Part One of my blog post I discussed the challenges as I see them posed by a BiModal IT approach to driving innovation and reducing risk. In this part of the post I’m going to talk about some of our experiences on the negative cultural impacts of Bimodal IT, and propose an alternative approach.

“Them and Us”

One of the biggest cultural impacts that we see with the Bi Modal IT approach is the development of a “them and us” type culture. While Gartner et al try to frame this in a positive light with terms like “sprinters and marathon runners”, inevitably it creates an environment where everyone wants to be in the Mode 2 “innovators” crowd, with Mode 1 engineers described in pejorative terms such as ‘reactionary’, ‘outdated’ or ‘blockers’.

Yet a huge amount of the knowledge and experience lies with the Mode 1 engineers and they form a key part of an integrated team. Often they are the people with the knowledge of the existing critical systems and processes, and the business they serve. They have a huge amount of organisational experience and know how to navigate the often complex, bureaucratic landscape of a large organisation. In short, they know how to get “get stuff done”.

Let’s also not forget that they often have a large amount of IT expertise and experience – even if they’re not necessarily currently using the latest and greatest DevOps tools. Often, when given the chance, they are more than up for the challenge of learning something new – and given the chronic skills shortage we have in the DevOps market, we’d be mad not to embrace that.

And at some point we, the Mode 2 team, will need to connect back into the “legacy” Mode 1 environment in order to make what has been developed actually relevant. If we create a “Them and Us” culture, what reaction do you think we’re going to get when we approach them for their help?

I think Bi Modal IT is less a strategy as such, and more a description of where we stand at present. It paints a picture of stalled IT transformation and it’s an idea that we need to quickly move beyond.

So is there an alternative?

It’s difficult to suggest an alternative to something that you don’t really consider to be a strategy as much as a description of failure, but if my main objections to BiModal IT are cultural, then the alternatives must themselves be cultural in nature.

If BiModal IT further entrenches silos within organisations then the strongest antidote to this is one which actively removes these barriers between people. The best example of this we have experienced to date is the multidisciplinary team used within product-focused delivery teams and pioneered by the likes of Amazon and Netflix.

This style of team works because

• It tackles the silo problem head-on
• It’s inexpensive and easy to implement – you can start as small as a single service
• It’s easy to scale as required – just add more teams (note: don’t scale the team, scale the workload)
• It goes way beyond just DevOps – the team can encompass whoever is needed to get the project done – legal, marketing, procurement; it doesn’t matter.

Summary

I have experienced first hand that Bi Modal IT is a counter-productive strategy for IT delivery. It fails to foster long term innovation, and at best is risk neutral for critical “Mode 1” services. Taking a multidisciplinary approach, moving forward together, I believe is the only way to go.

But isn’t transforming ourselves and moving together much harder? Of course it is, and I think that’s why BIModal IT has gained so much traction within the industry, it has the allure of the crash diet and the get rich quick scheme. But there are no shortcuts to transformation, only the daily struggle of a team pulling in the same direction.