Home Tech 5 steps to scale AI initiatives in the ‘new normal’

5 steps to scale AI initiatives in the ‘new normal’

by gbaf mag
21 views

By Andrew Duncan, Partner and UK CEO, Infosys Consulting

A top priority for all businesses in the age of COVID-19 is resiliency. AI is vital to achieve this, helping organisations and their employees better navigate uncertain supply and demand, adjust to disruptions in operations, and adapt to sharp changes in consumer priorities. In short, AI will help us predict what has often felt unpredictable this year. Some businesses will have already embarked on their digital transformation before the pandemic, giving them an edge at the start of the crisis. But, for many others, COVID-19 has been the motivation they needed to explore the benefits of AI. Whether they are well-versed in digital tools or have been reluctant to begin, most enterprises will face hurdles during the process of scaling AI initiatives from POCs to production. To achieve this successfully, leaders must look beyond the immediate horizon and put AI at the centre of their people, process and business strategies.

  1. Embed AI into long-term strategies

AI is a tool for solving business problems or achieving business outcomes. Any AI strategy should therefore be intrinsically linked to your COVID-19 recovery plans. Leaders must move away from thinking about how to apply AI in the short-term, and consider their boardroom’s long-term goals, focusing on solutions that explicitly deliver on elements of this strategy; for example, predicting customer buying patterns to drive hyper personalised products without sacrificing real-time supply and mass customisation. Executives also need to forecast their journey with AI to make sure they are investing in a programme that is simple enough to realise value from immediately and yet mature enough to grow with the enterprise as their needs for AI become more complex.

  1. Prioritise data governance

Every digital transformation journey, including scaling AI, relies on data at its core. However, given the rarity of a pandemic, the historical data that fed many of our analytical models quickly became out of date, incomplete and unsound at the start of 2020. In response, organisations must look to conduct audits to identify weaknesses and errors in their existing models and overhaul their data strategy. Underpinning this will be a new emphasis on data governance. But data governance doesn’t just mean more data; it means collecting, transforming and annotating the right data, both structured and unstructured. Knowing the desired outcomes of the AI initiative will be key in deciding which data may be more useful to collect and operationalise.

  1. Avoid tunnel-vision

The vast majority of blockers when scaling an AI project come not from the IT and delivery unit, but other areas of the business: slow budgeting processes, freezes on recruiting new roles, and lack of communication across the organisation. Taking a portfolio view of your AI programmes – looking at the business in totality and making decisions accordingly – will go a long way in overcoming these blockers. Rather than thinking of AI initiatives as a simple list of individual use cases, leaders should consider the collective success of these programmes over time. Shaping, iterating and investigating ideas holistically before a go/no go decision will make tracking ROI significantly easier in the long-term.

  1. Monitor for success

Transitioning from AI as a source of innovation to a driver of business value is something that many companies still struggle with. At the same time, AI programmes are often more complex than traditional software implementation projects. These factors mean that projects often fail to keep momentum and focus. Defining and tracking value throughout the course of an implementation, with emphasis on small but incremental gains, will help incentivise progress and keep programs on track. Measuring ROI can range from simple metrics like putting out a customer satisfaction survey, to the advanced use of machine learning to quantify responses to the implemented changes. Most importantly, there must be a strong alignment between the C-suite’s understanding of AI and what success looks like, and the implementation team itself.

  1. Take a human approach

COVID-19 has shown human-AI collaboration at its best. By scaling AI across their organisation, business leaders are investing as much in their people as they are in the technology itself. This doesn’t just mean establishing the right talent mix – although there will undoubtedly be a rise in demand for data scientists and engineers in 2021 and beyond. Crucially, it means embedding AI ownership and accountability into all teams, ensuring employees fully understand AI and how it relates to their roles. This will inevitably include a mindset shift to agile, as well as upskilling and reskilling people to be “data native”. We also recommend moving away from siloed units or departments to cross-functional teams – something I discussed in-depth in my last article.

It is now clear that the very fundamentals of traditional job roles and processes have been permanently changed by the pandemic, and AI will be vital to support these changes in the new normal, with many tasks and activities becoming automated or augmented. In turn, processes will become more efficient, and results optimised. Despite the challenges, organisations are sure to reap successful outcomes, if they put AI at the heart of their digital transformation strategy.

You may also like

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More