Ahead of this week’s Chief Data & Analytics Officers UK conference, Denodo’s Paul Moxon, Hitachi Vantara’s Gwyn Evans and Dataiku’s Nadim Antar give their top predictions for the UK’s data and analytics industry in 2020
More than 200 data and analytics leaders will get a taste of what the future has in store for them when they descend on London for Corinium’s Chief Data & Analytics Officers UK conference this week.
Featuring 60 speakers from innovative companies including Dyson, HSBC, Dixons Carphone, Sky and more, the two-day event will provide a unique window into how the industry will evolve over the next 12 months.
To uncover some of the key themes and issues that will be discussed, we asked industry experts from three of this year’s sponsors to share their top predictions for data and analytics in 2020.
A Shift in Data’s ‘Centre of Gravity’
Paul Moxon, SVP of Data Architectures and Chief Evangelist at data virtualisation specialist Denodo, sees multi-cloud data architectures, robotic process automation and data fabrics becoming more common this year
1. Hybrid Data Architectures Will Become the Norm
As cloud initiatives become more mature, with more and more data migrating to the cloud, the centre-of-gravity of an organisation's data will shift. The balance will tip towards data being spread across both cloud and on-premises data sources.
Similarly, data and analytics platforms will transition to a multi-location, multi-cloud architecture, increasing the complexity of securing, managing and accessing this data.
2. Robotic Process Automation Goes Mainstream
Robotic process automation (RPA) will become an accepted mainstream technology in most organisations.
This will result in increasing pressure on IT to deliver the data to drive the RPA workflows – typically through REST APIs. Organisations will look for data platforms that support multiple data delivery methods to support traditional analytics, machine learning (ML) or AI initiatives and RPA workflows.
3. The Use of Data Fabrics Will Increase
In 2020, the search for an effective integration platform to access and utilize data and deal with issues like data security and siloed information will continue. But increasingly, organisations will look towards using a data fabric to overcome these data challenges.
By combining historical and real-time datasets across multiple data silos, a data fabric offers a single, secure and consistent data management framework. It reduces data delivery times and supports the automation of data preparation and integration, enabling organisations to focus more on ML and AI initiatives.
Augmentation Takes the Crown from Automation
Gwyn Evans, Digital Insights Solutions Engineer at digital transformation specialist Hitachi Vantara, anticipates the rise of augmented intelligence, data contextualisation tools and machine learning platforms
1. Augmented Intelligence Use Cases on the Rise
I think 2020 will refocus attention on practical systems that augment human intelligence (to enhance creativity, reasoning and decision making) rather than systems that promise complete automation.
For those working in analytics, this should mean improved tooling to help profile, catalogue and explore data, as well as quickly prototype and test analytical pipelines.
2. Focus on Tools That Help to Contextualise Data
For data analysts and data scientists, the importance of problem context and domain expertise should never be underestimated.
In 2020, I’d expect to see progress in how industry records and re-uses domain knowledge to facilitate faster collaboration between data professionals and domain experts.
3. Systems to Store and Manage ML Models
Finally, it’s not just the sharing of data that is important to data professionals in an organisation. The sharing of analytical pipelines, analytical context and machine learning models is vital to reduce development time and repetition of effort for data scientists and analysts.
For 2020, I think we’ll see the emergence of platforms and databases that allow colleagues to better share and understand their approaches to analysis.
New Priorities for Europe’s Data Leaders
Nadim Antar, VP of UK and Northern Europe at software company Dataiku, thinks data leaders will prioritise updating their cloud data management strategies, transitioning to initiative-driven teams and making AI ‘explainable’ in 2020
1. Cloud Data Management Will Take Centre Stage
A great migration to the cloud is underway for businesses worldwide. But while there are fewer up-front IT costs associated with cloud platforms, they are generally more expensive overall, especially when resources are improperly managed.
Ultimately, companies that don’t actively develop a larger cloud strategy and manage cloud costs in 2020 will face an uphill battle to prove positive ROI with AI projects, racking up bills that aren’t necessarily offset by the financial gains or savings from the projects themselves.
2. We’ll See More Initiative-Driven Teams
In 2019, companies around the world created centralised teams to handle data initiatives and kick-start AI efforts.
These ‘centres for excellence’ give data teams the ability to get started and scale-up quickly while easily prioritising different projects. Yet, they also require quite a lot of discipline to ensure that the business is sufficiently involved in centre of excellence initiatives.
The trend for 2020 will be moving toward a more formalised way for data and business experts to work together: the ‘initiative-driven teams’ model.
This approach allows teams to be ‘spun up’ or ‘spun down’ based on the project for tight, expert focus, ensuring that the results align well with business needs and expectations.
3. Continued Focus on Explainable AI
Explainable AI, trust and bias were topics that took centre stage in 2019, but that does not mean they have now been resolved. What’s new for 2020 is that there is both internal and external pressure to make AI explainable, unbiased and trusted:
- Internal pressure comes from those designing AI systems as and employees who depend on them for their jobs
- External pressure comes from customers and end-users of products and services that AI systems affect
Ultimately, building internal trust will provide the foundation for external trust. This starts with trust in the data that is being used in AI systems. Data quality is one of the most basic but most important hurdles to overcome in the path to building sustainable AI that will bring business value.