More and more people understand that an increasing portion of life’s decisions are being made by algorithms. Whether we realise it or not, each of us experiences the influence of data-driven decisions every day in our work, shopping, and leisure. It may be selecting the next song on a Spotify playlist, which ad you’re served on a newspaper website, or it may mean the difference between the towns you chose to buy a house in.

The credit business recognised the importance of analytics in making better decisions just as consumers are better understanding how it brings them greater value and faster decisions in the app economy era. As this awareness and use increases, it is critical for businesses to maintain proper governance, transparency and management of the analytics decision-making process. Grappling with realising the benefits from algorithms in customer decisions is one of the major challenges for the credit industry, right now.

>See also: 3 ways technology drives online lending

Companies that use data and analytics to routinely make customer decisions can generate up to nine percent more revenue and be 26% more profitable than their competitors (Forbes “The Data-Driven Imperative: A Shift to Digitised Decisioning”). However, results like these can be difficult to achieve. Organisations face many challenges – accessing the right data, applying the actionable insights, governing and managing use of the data and analytics – to ensure appropriate treatment of customers. According to IDC research, only 15% of all insights created by analytics gets acted on in operational systems today.

Leveraging data and statistical models for decisions is not a new practice. It’s often seen in areas like consumer credit where the cost of getting things wrong is high and there is ample historical data available to make predictions. Furthermore, when these decisions need to be managed at a larger scale, specialised software has long been available to support the automation and governance.

Technology is driving a rapid expansion of data-driven decisions beyond consumer credit and into all types of industries and areas. Artificial intelligence accelerates this trend by making more information available (such as voice and video) to machines through computer vision and natural language processing. The ability to create model predictions from new types and larger quantities of data is being made much easier through machine learning and other advanced analytics techniques. The potential benefit to businesses is substantial. McKinsey predicts that AI may add a trillion dollars’ worth of additional value to areas such as management of risk, finance and IT over the next 20 years.

>See also: The customer journey: What it takes to add value in the digital era

Data, analytics and decisions are coming together

The convergence of data, analytics and decisioning technology is being is driven, not so much by new discoveries, but by a series of trends brought about through advancements in technology…

• Big Data technologies: Storing large volumes of data has become cost-efficient. This means many companies are starting to gather and hold vast amounts of data to use for further analysis. And, today, many organisations have a data-lake strategy to collect and store most of their data from operational systems and other sources.
• Artificial intelligence (AI) and machine learning (ML) AI and ML are removing the costs associated with the manual building of analytics to enable predictions and modelling that can be applied to a much wider range of problems and decisions. In addition, AI is enabling computers to automatically understand more complex information such as images and speech. This kind of automation is revolutionising a variety of tasks in all areas of business – particularly activities that humans would naturally find tedious or challenging to perform, such as correctly classifying the colour of a dress in an inventory or matching a person’s voice to a previous recording.
• Solutions to operationalise decisions: The technology to manage and operationalise decisions is becoming more widespread across the industry. By using tools that ensure the right model is used, that validate model performance and that monitor and explain how a model is making decisions, companies can now insert automated data-driven decisions at key points in their processes.

>See also: How mobile has transformed money lending

• The digitisation of the world: As consumers adopt more digital technologies and the Internet of Things continues to grow, the amount of data available to businesses also increases. This data can be used to improve customer interactions and the decisions businesses make when managing their customer relationships.
• Blending business understanding with data-science: Technology and business leaders recognise the need to bring together data-science and business expertise and acumen to get a better understanding of the data and the insights that can be drawn.
• Access to insights with business control: The challenge to ensure that the predictions and insights created by the algorithms are properly understood, monitored and controlled. Depending on the type of decisions, the cost of an error in an algorithm can be significant, as has been demonstrated in a variety of settings, such as flash-trading or autonomous vehicles.

Where the business understands and integrates the right mix of solutions the decision-making process can be extraordinarily streamlined. Getting there is a highly consultative process involving listening to customers, the needs of the business, and to what the current IT landscape tells is telling you.

By Boris Huard, UK&I Managing Director of Decision Analytics, Experian