Creating an AI-enabled organisation: moving from analytics to operational impact

AI can become a major driver of productivity and business success. But it requires a significant reframing in how data science is perceived and used, moving from analytical insights to operational outcomes. This shift is as much cultural as organisational, but it’s crucial to becoming an AI-enabled organisation.

Increasingly at Faculty, we find ourselves meeting with Chief Data and Analytics Officers (CDAOs). Typically this is a relatively new role, introduced into our customer’s business in the past year, undoubtedly influenced by recent developments in artificial intelligence (AI) and the resulting increase in awareness and strategic focus in this field. Their task is to get value from data.

But what does it mean to ‘get more value’ from data? In the past, it used to mean pumping more data into analysis and allowing insights to flow through to better decision-making. But it turns out that approach has not fared well, with 65% of businesses believing that decision-making has become more complicated, despite having access to more data than ever before.

Analysis paralysis - a brief history

Let’s go back 20 years when the role of data scientist was barely a glint in the Chief Technology Officer’s eye. Indeed, the CTO role itself was still nascent or non-existent in many companies. This was a world where business leaders were being cajoled by consultants into becoming ‘technology companies’, while executives clutched their pearls in response, horrified by the prospect.

In this world, the business analyst sat at the vanguard of data analytics, nobly producing metrics and KPI dashboards and wowing Chief Financial Officers (CFOs) with their slick charts and management reports. The cutting edge was a moving average forecast that magically predicted the future from an otherwise indecipherable mass of numbers.

It feels like a long time ago. Yet, despite rapid technological advances, the way data is used in most businesses has changed surprisingly little. Many companies now count data scientists amongst their clan. But while they offer a more advanced analytical skill set and the ability to code, in most cases their roles are an evolution of the suited business analysts of yesteryear. They often sit within business units, restricted to producing static analyses and ‘insights’, without ever realising their ambitions of developing algorithmically driven systems to really help the business perform better. Meanwhile, managers are left puzzled as to why the seemingly interesting analysis produced doesn’t yield a more meaningful business impact. 

The reason for the stalemate is precisely this focus on analytics. The reality is that business units, under pressure and fast-moving, want to solve problems quickly and make better decisions but they lack the culture or know-how to develop the tooling to do this sustainably. Instead, they are trapped in a vortex of producing stand-alone pieces of analysis that don’t connect with each other or integrate into business processes. At best, these provide a brief moment of clarity before becoming obsolete. At worst, they produce a costly and inefficient cycle of analysis that is doomed to fail.

The disconnect between data scientists and IT teams

Compounding this is the fact that software tooling is often viewed as something separate from the core business.  Something that other people inside your business procure off the shelf from a third-party developer to perform all manner of things, other than analytics. 

So while most successful businesses have achieved their destiny of becoming technology-driven companies, the vast IT teams created rarely interact with their data scientists to make use of the data they curate.  As a result, there is an enormous missed opportunity to connect the dots between these two disciplines to deliver real business value.

AI as an operational discipline

Developments in AI bring this situation into sharp contrast. AI unlocks the ability to make use of large, complex data for productive operational purposes through software tooling. 

Analytics is primarily backwards-looking, providing information to help monitor and understand the past through graphs and KPIs on slides in meetings.

In contrast, AI-driven software is forward-looking, putting levers into the hands of the operational business to help them understand situations and anticipate the future impact of actions in real time. It sits inside operational workflows, helping them run more smoothly and efficiently.

AI is an operational discipline rather than an analytical one. Using it to create cleverer charts misses the point and is doomed to underwhelm. So when data science grows out of analytical teams there is a risk this heritage creates the wrong focal point.

Becoming an AI-enabled business

Creating an AI-enabled business requires a different approach. One in which data science is built into multi-disciplinary product teams focused on injecting AI into the core operational workflows of the business. 

These product teams need to integrate data science and software technology capabilities and be empowered to work closely with the business to improve how it works.

This is no easy task. Data scientists are typically dispersed throughout the business for good reason, as it allows them to gain crucial business understanding. Software technology capability often exists only in the IT team, with little culture of applying these skills to internal solution development. And business users are busy and often hard to engage with, but their input is essential to developing useful operational solutions.

Bringing these capabilities together successfully requires navigating operational, political and cultural barriers in the organisation. But we’ve been here before, with the digital transformation. While there are no easy paths to success, businesses that move quickly and decisively in this direction will reap the benefits and set themselves up for future success.

With a decade of experience successfully developing AI solutions for customers, at Faculty we are well-versed in navigating and overcoming these challenges. We not only develop AI-driven operational solutions for our customers but, crucially, also ensure they are able to grow this capability for themselves.

Successful businesses of the future will be ‘AI-enabled’. They will put AI at the core of their operational processes. And this will be the hallmark of a successful CDAO.

Are you ready to shift focus from analytics to operations and make your business an AI-enabled powerhouse? Let’s start a conversation.