Revium’s Data & Analytics team recently attended the Melbourne CDAO conference where the latest advancements in Data, Machine Learning (ML) and Artificial Intelligence (AI) were presented by industry leaders.
High profile keynote speakers were featured from a range of sectors including:
- Government (ATO and BOM)
- Energy (AusNet)
- Finance (GE and Bank West)
- Insurance (QBE and IAG)
- Technology (Oracle and NARTA)
- Not for Profits (Heart Foundation)
- Education (Monash Uni, NZ Tertiary Education Commission)
While all speakers presented on unique topics, clear themes appeared throughout all presentations:
- Data, ML and AI all have their challenges
- Buy-in from senior leadership is essential
- Expect the first project to fail
- Democratisation of data and AI is required in the business
- The ROI is strong once you have pushed through the initial teething stages.
Challenges of Data, ML and AI
There are over 27 billion active devices connected to the internet. Every second 127 new devices are added. This highlights the sheer volume of data that is available to work.
And it’s intimidating.
4 years ago, 37% of Chief Data Analytics Officers (CDAO) believed they were working for a data-driven organisation, but this figure has dropped to 31% today. Despite the fact that there is more data, CDAOs don’t believe that they are doing enough with what is available, so they don’t consider their organisations as being driven by data.
Knowing where to begin when working with this ever-growing data can often be the biggest challenge an organisation can face. Businesses need to invest in new ways of managing data and then develop new ways to leverage it in order to empower them to make truly informed decisions that create positive business outcomes.
An increasing amount of data leads to an increase in siloing as each business unit collects and processes customer data via their own unique ways of working. From customer service to accounts and sales, each team will have their own method of data collection and processing. Streamlining these processes across business units into a centralised data store provides a significantly improved understanding of a customer journey than trying to draw insights from each individual part.
Bringing together this data to form a usable set creates its own challenges, as staff on the ground need to learn new ways of working that reflect a data-driven culture. Businesses will need to focus on developing this culture and communicate this often and clearly to employees.
Senior Leadership Buy-in
The power of predictive analytics and ML in solving business challenges is at the fore-front when it comes to reasons for adoption, and highlights the need for senior leadership to get on board. Bain & Company suggest that companies that utilise ML and predictive analytics are more financially successful.
Origin Energy has also embraced data in an attempt to address low customer trust and stem customer attrition. By abandoning more traditional, slower data collection methods Origin was able to use data to change processes and meet their business goals.
To get the buy in from senior leadership, focus on streamlining process and how you interact with your customers. Provide a proof of concept before trying to blow them away with disruption and innovation. This can be done by focusing on the data that you already have and utilising ML and AI to demonstrate the outcome.
Starting from simpler and more easily accessible places allows businesses to create a ‘north star goal’, get off the ground and then work through a maturity growth curve towards the end goal.
Expect the First Project to Fail
AI and ML follow a lifecycle that improves with time, so setting stakeholder expectations from the outset is critical. Educate and train stakeholders to help them understand how AI and ML work in the context of the business challenge they are trying to solve. Starting small with a solid concept builds confidence in the data and allows for human oversight to guide and develop the output.
Democratise Data and AI
Control over data is necessary in any organisation but it is also critical to promote information and technology across the business. Focusing on the benefits that come from adoption will remove the stigma around technologies like AI so that staff can focus on the benefits instead.
By doing this, staff are enabled to better understand its how it can improve the business for them and their customers. Highlight examples like sentiment analysis (which can let staff know ahead of time what their customer is thinking and feeling). Staff can then enter conversations with customers armed with the right solutions to solve their problems.
AI is often perceived as a job killer, but instead it should be framed as a tool that can help give staff time back to do the interesting work they would prefer to do instead of the bland repetitive tasks that AI can manage.
Revium is constantly in conversations with our customers around all of these issues - helping them to overcome the same challenges and supporting them in developing solutions specific to their business. This process is a journey and it can be difficult at times, but it is one with highly valuable outputs. We recommend you start small, work within an existing framework to prove a concept that can be scaled up, take learnings from each step and apply them for future iterations. Make sure you share the journey with teams across the business to create a network of advocates for the program that will help to ensure business wide adoption and success in the long term.