Architecture driving innovation
What makes Architecture valuable?
We all value Architecture differently. Our assessments of its significance — be it the Sydney Opera House or a payroll system — are based on our own frame of reference and personal experience of ‘what works best.’
Recently I presented at the Enterprise Architecture Summit in Melbourne and gave the example of the Notre-Dame cathedral.
It is recognised as an architectural masterpiece, but it took over 300 years to build and was certainly not delivered on time or within budget. While the return on investment has been good over the (very) long term, the design was set and couldn’t be changed for hundreds of years, even while better ideas emerged.
Compare this with building a hut. They are cheap and quick to build and use readily available skills and materials. They are fit for purpose and cheap to maintain. The consequences of making a mistake and starting again if it doesn’t fit in with the landscape or impacts other properties are low. You can rip it out and start again.
This is the fundamental difference between Architecture in the pre- and post-cloud world.
Pre-cloud and post-cloud
Before the cloud made change quick, cheap and easy, IT was a constant. Around 5% of programs finished on time or within budget because programs were massive and relied on long term delivery timeframes with armies of resources. Enterprise Architects often spent a long time making decisions because of the potential impact. When IT failed, it did so slowly and at great cost.
In the Amazon Web Services (AWS) and Software as a Service (SaaS) world, Enterprise Architects must start to redefine their value. Today, many EAs are spending more time translating business requirements into SaaS products that we enhance and customise rather than developing solutions to build and deploy.
This shift is driven by CIOs looking for innovation from tech teams: focussing on disruption and incubation while managing their digital footprint.
Reimagining the Architecture practice at Transurban
Transurban may be small compared to other companies in the ASX Top 15 with only 2,000 employees, but we make a significant investment each year in our technology capability.
Does Transurban sit in the new world of the old? For us, it’s not really a question of one or the other, but how we can deliver value in both.
To do this, we operate what Gartner call Bimodal IT. This means carefully managing large scale, high risk projects using traditional waterfall methods; while at the same time, embedding architects in DevOps teams encouraging them to iterate and fail fast.
To succeed in both means transitioning from a highly centralised architecture function to a distributed model. We have Enterprise and Program Architects sitting across business platforms and programs of work, while Solution Architects work within a given technology domain. All contribute to the broader governance framework and form part of our Architecture Community of Practice.
Architects are given the support they need; have the mandate to make key decisions, and are sponsored to try things on their own terms.
This approach keeps our best people engaged and empowered to deliver on existing patterns and removes some of the barriers created by hierarchical decision making. Architects are given the support they need; have the mandate to make key decisions, and are sponsored to try things on their own terms. It’s important to note that this isn’t a special project set up outside Technology, and that Transurban is investing in both bimodal streams.
Looking to the future, we see a place for both modes in the long term. When the Agile/DevOps world eventually become the mainstream (as many predict it will), Transurban will have the skills and experience to respond.
