In my previous post The Role of Artificial Intelligence and Machine Learning in Digital Transformation, I provided an introduction to Machine Learning (ML) and Artificial Intelligence (AI). In this post, we will build on that knowledge and take a look at why they are necessary capabilities in order for an organisation to reach a mature digital state.
Let’s start by considering what I mean by a mature digital state. Probably the most commonly used models to assess maturity, are those published by Gartner. Each maturity model shares a common framework and provides an approach to assess a given capability on a scale of one to five across a number of domains. For example, In the case of Gartner’s Government Digital Maturity Model, the five levels of maturity are:
- Level 1 (Initial) E-government
- Level 2 (Opportunistic) Open
- Level 3 (Repeatable) Data Centric
- Level 4 (Managed) Fully Digital
- Level 5 (Optimising) Smart
While I do not have access to Gartner’s Digital Maturity Model research, I would expect it to include five or six dimensions to grade the organisation’s maturity against. One of these dimensions will almost certainly be Data, with possibly sub-categories of data types, business process data, product & service data and customer data.
In order to argue about what’s necessary to attain the more mature states of Managed (Fully Digital) and Optimising (Smart), we need to understand, at least at a high-level, what an organisation would look like in each of these states.
In the article, When Less Becomes More: The Journey to Digital Government, Fully Digital is summarised by
the organisation is completely data driven and is regularly exploring opportunities to innovate by combining together data from different departments. The organisation not only sees data as a strategic asset, but is effectively gaining insight from it. Traditional and new services are provided through a variety of channels.
If we now consider how an organisation could be considered as Fully Digital within the sub-dimension of products and services, one characteristic could be:
- the organisation uses business rules within a recommendation engine to proactively drive customer purchasing behaviour
another example could be:
- the organisation has a data architecture that allows for assets to be effectively shared across the entire business
The first attribute places us in the domain of expert systems, a sub-field of AI. The second attribute talks to form or structure, a possible topic for a future article. While an organisation at this level will possess intelligent systems, there’s no measure of how smart those systems are or whether there is a feedback mechanism in place to improve the performance of those systems over time. Such a feedback or learning mechanism takes the organisation into a continually optimising state or Smart state as per the Gartner terminology.
Let’s once again refer back to the article, When Less Becomes More, this time to look at how a Smart state is characterised:
Not only is the business completely data driven, but the process of how the organisation leverages data has matured to the point where it is repeatable and predictable. Constant improvement and optimisation are the focus.
There is only one practical way for constant improvement and optimisation across an entire organisation, well one that is economically viable at least; it is to employ on-line ML systems that constantly evolve based on inputs and outputs with the goal of optimising some desired outcome. For example, if we consider a recommendation engine, one complimentary system could be an ML system that is optimised on conversion rates (impressions to clicks) where the only inputs are the location of the recommendation within a browser and user agent (desktop, mobile browser). Through a series of A/B testing (moving the location of the recommendation) and measuring the performance (conversion), the system can evolve and constantly be in a state of optimisation.
Every organisation that has Digital Transformation as a strategic focus needs to develop a strong capability in the areas of AI and ML. Initially, this capability may come through partnering with external teams with a proven track record. However, if data is truly a strategic asset, there’s a strong case for larger organisations to build an internal data science and platform team to strengthen and sustain their competitive advantage.
Connect with Red Hat Services
Learn more about Red Hat Consulting
Learn more about Red Hat Training
Learn more about Red Hat Certification
Subscribe to the Training Newsletter
Follow Red Hat Services on Twitter
Follow Red Hat Open Innovation Labs on Twitter
Like Red Hat Services on Facebook
Watch Red Hat Training videos on YouTube
Follow Red Hat Certified Professionals on LinkedIn