Achieving Digital Innovation through a Technology Centre of Excellence (COE): Learnings from an experienced practitioner

One of the things I talk about in my Digital Strategy course that I teach at the Australian Graduate School of Management is that developing a successful Digital Strategy is never an end state but an ongoing process of "Digital Evolution".

Organisations that are successful at using Digital technologies to create competitive advantage succeed by building processes, organisational structures and systems to constantly evaluate their business landscape against a constantly evolving technology landscape to identify new opportunities. Those processes, organisational structures and systems allow these organisations to select and solve the right problems at the right time with the right technology.

In this way, a successful Digital Strategy is not something you create and then forget, rather it evolves and changes as the business changes and technology changes. Whilst Business Strategies are built to last years, Digital Strategies rarely last more than 12 to 18 months before they need to evolve and change given the current pace of technological change. Digitally backed innovation becomes a critical part of an organisation's Digital Strategy and potentially a key Competitive Advantage. But how do you build Digital Innovation into your organisational DNA? One way organisations have been trying to do this is by setting up "Centres of Excellence" or COEs. Artificial Intelligence COEs, Automation COEs, Cloud COEs, we tend to hear every potential version of the COE concept these days some more successful than others. But is this the only way you can innovate and what is the best way to set up a COE if you decide to go down this path?

In this month's "Thoughts on Digital Strategy" article I talk about the process of Digital Innovation within organisations and the different structures and processes that organisations have used to build Digital Innovation into their DNA. I also give you some practical insights into how to incorporate Digital Innovation into your organisation by interviewing Stephen Blunt from the Automation and Innovation Hub (AIH) at the University of Sydney. The AIH is a COE centred around new technologies such as Robotic Process Automation (RPA) and Artificial Intelligence (AI). Stephen set up the AIH around 3 years ago for the University and it has been hugely successfull. The AIH is an example of what business strategists would call the "Specialist Mode" for organisational Digital Innovation. Stephen has a lot of practical experience around the challenges and opportunities associated with developing the COE concept and more broadly Digital Innovation in general. So his answers to my questions should prove interesting to those that are considering deploying a COE into their organisation.

But first, let's talk about innovation (and technology innovation) as a process to provide a scaffold for my subsequent discussion with Steve...

Innovation as a Process

Innovation theory and practical experience tell us that being innovative is not just about coming up with a good solution to a problem, rather it is as much about picking the right problem to solve from a number of potential problems, and then being able to properly articulate that problem in a way that allows you to pick the right solution from a number of potential solutions. The diagram above illustrates the process holistically. In summary, organisations need processes to ...

1. Generate new ideas in terms of problems that can be solved;
2. Prioritise and select the problems worth solving;
3. Identifying new ideas in terms of potential solutions to these problems (from a business and technology perspective); and then
4. Pick the right solutions to invest in

All you need is for one of these innovation sub-processes above to be problematic and the whole technology innovation process fails. Organisations often fail in being innovative by focusing on one or two steps in this process and not designing their organisational processes and structures to support all four steps.

Both the process for identifying problems to solve and the process for identifying potential solutions to those problems relies heavily on an organisations ability to generate good ideas in both of these categories in the first place. So let's talk also about the idea generation as a subcomponent of the Innovation Process...

Idea Generation as a Process

Idea Generation like many business processes is stochastic in nature. There will be some good ideas, some bad ideas and times when you have a large volume of ideas and times when you don't. Ultimately your goal is to generate the largest number of high-quality ideas both in terms of potential problems to solve and also in terms of potential technology solutions.

Trying to apply a more scientific view to this concept of idea generation, if you were to plot the quantity vs. quality of ideas that your organisation generates onto a chart you would likely get the bell-shaped (or "Normal") curve found in most statistics textbooks and shown in my diagram to the left. As previously mentioned, organisations, generally speaking, are trying to increase the number of high-quality ideas from which to choose from both in terms of problems and solutions to those problems. This is represented by the shaded area on the curve above. There are a number of strategies you can employ to increase the number of higher quality ideas.

For example, you can try to simply generate more ideas ( and therefore in proportion more higher quality ideas) or shift the curve to the right. You can use tactics like providing employee financial incentives or fun programs such as hackathons to increase the sheer volume of new ideas. But if you do this, you then need processes that can filter the better ideas from the large volume of poorer ones and your selection costs can go up leaving less money to actually implement these new ideas. Some organisations will then try to reduce these select costs by putting in processes that allow the people who are generating the list of ideas to self-select and pick the best ones but these processes often have their own challenges.

Another option is to improve the average quality of ideas by shifting the entire curve up. Some organisations do this by bringing in design and technical solution experts and or consultants to support idea generation and solution development and deployment processes. But research shows that this actually has the effect of flattening the curve whilst shifting the curve up which can mean you get more really good ideas but also some really poor ones as well. We have all heard stories of projects that high priced consultants have sold and delivered that have been colossal failures. So again you need good selection processes and even better risk mitigation strategies.

Ultimately there is no right or wrong answer in terms of how to design an effective Innovation Process or the supporting Ideas Generation process into your organisation. A lot of what will work will depend on the organisation itself, it's culture and its history. But you do need to have a think about these processes and design them into your organisations DNA if you want to promote effective Digital Innovation. The other thing you need to consider is how to design your organisation to support Digital Innovation in general. There are several ways you can design innovation into your business organisation structure. Business school theory tells us you have four options...

The Four Modes of Organisational Innovation

Consultants and academics alike love a 2 x 2 matrix. Being both a consultant and an academic I am no different. In the context of organisational innovation, HBR creates an effective 2x2 descriptive matrix by considering organisational design for innovation in terms of Organisation Scope (Internal/ External) vs. Incentive Mechanisms (Formal/ Informal.

What this framework says is that you can have a structured or formal approach to organisational design for innovation (sometimes called Closed Innovation) or you can have an informal (sometimes called Open) organisational design for innovation. Whether you prefer a formal or informal approach you also have to decide whether the innovation will come from within your organisation or outside of it, which is the other part of the 2x2 matrix.

So for example you can assume that your innovation will be generated from inside your organisation but come from across the entire organisation. Called the "Venture" mode, organisations that operate in this mode encourage all their employees to come up with new ideas and have processes and systems that help their employees pick the good ideas and give them financial backing to support and further develop those ideas. An example of this mode is when Google encourages all its employees to take 20% of their time to solve big problems for the company with no immediate benefit. But this mode often requires a good corporate culture of entrepreneurialism to be successful.

Keeping innovation internal, some organisations will also set up special groups or teams to drive innovation. Often this is done when the organisation's culture will not support open innovation processes. This is called the "Specialist" mode of organisational innovation design. Another name for this is a COE. In the next section, I will give you a practical example of this in the AIH at the University of System.

But not all innovation comes from within an organisation. For example, some organisations like Microsoft have set up formal partner networks to sponsor innovation from the outside. This is called the "Network" mode of organisational innovation design. In the Microsoft case they actively encourage their partners to build solutions on top of their technologies to address gaps in the technology and/or to develop new market opportunities. This creates a structure process for generating innovation from outside of the organisation.

Finally, organisations may still rely on innovation to come from outside the organisation but may adopt a more informal approach. For example by sponsoring user groups, conferences, etc. to foster the free flow of ideas around their products or services. This is referred to as the "Community" mode of organisational innovation.

Again there is no right or wrong answer here in terms of which mode to employ. Many organisation employ multiple modes and structure their organisation accordingly. Often the right answer depends on the industry the organisation is, its culture and its history.

Now let's talk about a specific case study of Digital Innovation, the Automation and Innovation Hub at the University of Sydney...

A Practical Example of the Specialist Mode of Innovation (often commonly called the COE) at the University of Sydney - The Automation and Innovation Hub

The Automation and Innovation Hub (AIH) at the University of Sydney was started by Steve Blunt back in 2018 and has become one of the largest RPA and AI COEs in Australia having won a number of awards for the work that they have done. So I thought my readers would benefit from Steve's insights into the Innovation Process. The following is the text of my interview with Steve a few weeks back...

Q: So Steve, what made you start the AIH at the University of Sydney?

 A: After considering the University’s challenges, it became clear that no one had found a way to fix problems in a way which was sustainable. Automation was just starting to gain traction, and it seemed like a great opportunity for the University be in the forefront. It also felt like there were endless consultants with large budgets trying but ultimately not successful at improving service levels - so why not build an internal capability instead? The Automation COE idea held promise because we would be able get automation and AI solutions quickly without having the large overhead cost or risk involved of a large scale project, while leveraging what's already happening within financial services sector as a guide!

 Q: What have been the advantages of having a centralised innovation team at the University?

 A: The best thing about having a centralised innovation team is that we can build and share knowledge among ourselves, so the team is always getting better. This makes automation more efficient, enabling us to reuse assets versus individuals trying it on their own, likely without support. Plus as a dedicated function, there isn't any competition for your time either - automation always takes priority here at AIH so there isn’t anything holding the team back from delivering new solutions quickly.

 Q: What have been the challenges of running an automation COE?

 A: The University has come a long way and is very automation friendly, but it's been an ongoing journey to get here. The number one challenge for us in running these kinds of projects has been managing all types of people issues - RPA and AI technology is proven and reliable - but navigating internal politics, roadblocks and competing priorities can consume substancial amounts of time.

Q: What was your processes for generating new project ideas?

A: To come up with new project ideas, I talk to people. What I tease out are their key priorities, pain points and things they'd love to offload. This insight into their key drivers and pain points really gives us the long list of problems worth exploring. Then we look at size and potential impact from resolution, readiness of the business, and come up with a pretty reliable and deep list.

 Q: What was your process for picking the right projects to work on?

 A: We take our long list and evaluate the ideas based on size of the problem, potential impact from resolution, and readiness of the business. My clever team then brainstorms potential solutions to the problems, so we understand how feasible and level of effort required before investing any time. If the problem is worth solving, the impact is meaningful, the business is keen, and the solution fits within our remit (automation and AI), then we're off to building the solution! We will often break a problem down into smaller chunks with a view to getting a solution into the hands of the business more quickly and being more agile.

 Q: What was your process for identifying solutions and picking the right solutions to a problem?

 A: We examine the current process, and look at if this problem has been solved before (is there a solution in the market), spit-ball potential ways to solve the problem and debate the merits amongst the team. If there are multiple potential solutions without a clear winner, we rapid prototype and A/B test to see which makes sense.

 Q: What advice would you have for other organisations setting up their own Centres of Excellence like the AIH?

 A: Start by being super clear on your purpose and remit. Automation and AI can be quite broad, so understanding what's in your wheelhouse and what's not, helps set the tone, provides clarity of your service(s) and reduces potential clashes with other trying to do the same thing.

Conclusions

There are many ways to build innovation into your organisation. One of these ways is the Innovation Centre of Excellence or COE. But it is not the only way and organisations need to carefully consider what is the best mode for them. Innovation is a process with many parts and organisations need to make sure they include all those parts into their organisational design. Finally idea generation both in terms of defining problems to solve and coming up with solutions to those problems is an absolutely critical part of the innovation process and organisation need to consider carefully what their processes and strategies will be in this regard.

Dr David Goad has a number of roles including Industry Professor for Deakin University, Adjunct Faculty for AGSM (UNSW), Honorary Associate for the University of Sydney, CTO for IBM Consulting Australia and New Zealand and is a Regional Director for Microsoft. He frequently is asked to speak at conferences on the topics of AI, IoT, Cloud and Robotic Process Automation and teaches Digital Strategy, Digital Transformation and Executive Education Courses for a number of universities.