Share with your friends
Confident decision making

Predictive analytics, it works

Predictive analytics, it works

Data analytics is a major function of any modern business. Regardless of industry, companies need data to evaluate their operations and plan future growth, making data analytics and advanced techniques, like predictive analytics, an essential business tool.

Canadian CEOs recognize the value that predictive analytics delivers to their decision-making process. In fact, almost every meeting we attend with Canadian CEOs inevitably turns to the subject of predictive analytics and new opportunities for it to deliver more.

There is a reason CEOs want to talk to us about predictive analytics; it's because they know it is the key to vastly improved decision-making.

However, what can Canadian CEOs do to enhance their data analytics capabilities and improve decision-making? Here are five ideas based on our experience working with leading organizations in the US and Canada:

  1. Understand what predictive can do for you. Take a more aggressive approach to exploring predictive analytics and spend time talking to advisors, startups, incubators, and service providers about what they are seeing in the market. Assess where predictive analytics and insights might improve your business value and begin testing as soon as possible.
  2. Embed analytics into your culture. Develop a practical and strategic roadmap for digitization and analytics. Find ways to encourage employees to integrate analytics into their work processes. Remove data siloes and scale up analytics successes. Move the predictive analytics team down the hall from the executive suite to provide immediate, unfiltered access to insights.
  3. Focus on small yet reliable data sets. Massive data sets are great, but a lot of valuable insights can also be uncovered from small, high quality data sets. Instead of spending all of your time and resources on gathering as much data as possible, start exploring what the data could be telling you and then focus your efforts on finding and curating the right data to support deeper insights.
  4. Grow your in-house capabilities. Consider what new capabilities your organization will need in order to effectively operate in a data and analytics-enabled environment. Develop both transitional capabilities as well as core capabilities to facilitate transformation and continuous improvement. Reassess your capability requirements frequently.
  5. Test your models thoroughly.  Nobody is suggesting you hand all decision-making over to the machines without first testing them using historic data sets and real-life scenarios. The higher the risk potential, the more testing should be conducted before going live. Assign ownership of the algorithm to an individual to ensure it continues to operate as expected.
  6. Operationalize predictive analytics into everyday decision making. One-time proof of concepts and insights generation are a good place to start, but often such initiatives end after the initial buzz. Have a strategy and an execution plan to evolve concepts and pilots into production grade, prediction driven, intelligent automation systems. Direct your architecture design towards application programming interfaces (API), micro-services and containers in a platform environment which will allow your organization to snap decision and workflow modules together, Lego2 style, to rapidly build and deploy intelligent applications.
  7. Don't forget the data foundation. Predictions are only as good as the data fed into models during both the design and operational phases. The need for good data is much greater now than ever before and your organization needs to have a robust enterprise data strategy in order to deploy intelligent automation at scale. Firms that recognize the value of good data as a key enabler and take appropriate action, will leap ahead of their peers.

As a continuously evolving field, data analytics enables companies to understand their clientele, operations, and market more deeply by spotting trends, understanding the performance of products / services, and their market position. Successful businesses are those that leverage and push the boundaries of analytics to attain reliable, predictive insights, unleashing opportunities for operational performance, risk management, and improved customer experience.

For example, energy businesses are using predictive analytics to launch maintenance requests using live data. In the retail sector, predictive analytics is being used to identify the best locations for future stores and reimagine in-store customer experience. Financial Services businesses take advantage of predictive analytics by helping compliance officers create new models for managing risk.

Companies that use predictive analytics win far more often. According to a report from the Society of Actuaries, healthcare businesses that use predictive analytics save around 15 percent on their annual budgets and can attain savings of over 25 percent over a 5 year period.[1] For instance, healthcare organizations deploying predictive analytics in their operations have reduced hospital readmissions, forecasted usage, improved their supply chain, managed staff more efficiently, and offered better care at a lower price.

Over the past few years, many companies have adopted systems that utilize predictive analytics to save energy costs. By automatically adjusting lighting or temperature, businesses can save over 90 percent of their energy costs.

We have a prediction: within the next three years, all corporate decision-making will be influenced, in some way or another, by predictive analytics.

*All statistics result from the 2018 Canadian CEO Survey.

[1] Society of Actuaries February 2017 Report, "Predictive Analytics in Healthcare Trend Forecast."

[2] Lego is a registered trademark of the Lego Corporation