Let’s say you decide to build a new house. Not only do you have to buy the materials, but you also have to hire the skilled talent who can get the job done. That is a lesson many CIOs around the world are learning about their plans to implement machine learning technologies that are able to analyze and improve performance without direct human intervention. Despite investing in machine learning, a new survey from ServiceNow indicates that most CIOs do not have the talent, data quality and budgets to fully leverage the technology. If your organization is embarking on the machine learning journey (and it should be), there are five steps CIOs must take to maximize the value of their investment.

Take these steps today, because the long-awaited Age of Machine Learning may be upon us soon. Computer science has caught up to the hype around machines that emulate human intelligence. Now, the technology occupies the peak position of Gartner’s Hype Cycle for Emerging Technologies, indicating that it has matured enough to spur wide interest. In other words, your competitors are also investing in machine learning.

Five hundred CIOs were recently polled for the annual Global CIO Point of View Survey, and the findings reveal that businesses are preparing for the widespread adoption of this transformational technology to automate decision making. Nearly 90% are using machine learning in some capacity, and most are still developing strategies or piloting the technology. However, the full potential of machine learning remains largely unrealized. For most organizations, many decisions still require human input. Only 8% of respondents say their use of machine learning is substantially or highly developed, as opposed to 35% for the internet of things or 65% for analytics.

Designing an organizational structure to support data and analytics activities, an effective technology infrastructure and ensuring senior management is involved are the three most significant challenges to attaining data and analytics objectives related to machine learning, according to a McKinsey study. It goes on to claim that organizations that can harness these capabilities effectively will be able to create significant value and differentiate themselves, while others will find themselves increasingly at a disadvantage.

Capturing greater value requires more than investing in technology. It is also necessary to make significant organizational and process changes, including approaches to talent, IT management and risk management. Making progress requires following five steps:

1. Improve Data Quality

Ensuring the quality of data is a common obstacle to machine learning adoption. Poor data leads to machines making poor decisions, which can lead to increased risk. CIOs need to consider implementing solutions that simplify data maintenance in order to accelerate the transition to machine learning. The first step should be to consolidate redundant legacy and on-premise IT tools into a single data model.

2. Establish Value Realization

Articulate the business value of all technology goals, then determine how best to reach those goals. This includes examining existing processes to identify which unstructured work patterns will benefit most from automation. Determining where fragmented data “lives” will enable you to identify how automation will lead to gains in productivity.