The Future of AI and ROI for the Enterprise

Scaling AI Joy Looney

For many years, AI was an experimental risk for companies. Today, AI is not a brand new concept and most enterprises have at least explored AI implementation. As of 2020, 68% of enterprises had used AI, having already adopted AI applications or introduced AI on some level into their business processes. So, where does AI stand moving forward? Recently, Dataiku spoke with Mike Gualtieri, VP & Principal Analyst at Forrester, in “The Future of AI and ROI for the Enterprise, featuring Forrester” webinar about the current state of the market and what AI success looks like going forward. Let’s explore some of the key takeaways from the conversation!

A Lot to Gain and Room to Grow 

AI may not be a shiny new thing anymore, but that doesn’t mean that it is any less enticing for the future of the enterprise. The results shown from companies who are now in the process of maturing their AI applications are incredible. Forrester conducts surveys to measure the impact of AI and the feedback received has revealed overwhelmingly positive reactions to AI applications. The ROI that companies have been able to achieve is nothing short of impressive. In fact, a single use case (if executed with the right combination of people, processes, and technology) can reap millions of dollars in benefits. As AI is used for more and more use cases across multiple scenarios and industries, the ROI companies see is expected to increase. 

The AI software market will exceed $36 billion by 2025, even as it gets absorbed into the broader software market.” - Mike Gualitieri, VP and Principal Analyst, Forrester

An overarching view of the AI market reveals that AI has reached inflection — meaning, someday, in the future, every single enterprise is going to use AI. Hundreds to thousands of use case opportunities are untapped and ready for AI implementation. For the future, AI is an essential capability, much as other software was in the past and present. Aware of this rapidly approaching reality, business leaders have started to focus on the best execution tactics for AI implementation moving forward. Integrating business needs and goals with AI will enable organizations to stay fully informed on the happenings of customers, competitors, and the market in general. 

From Experimentation to Implementation 

Increasingly, the conversation among enterprises is shifting from experimentation with AI to discovering the best way to fully evolve an organization to support and derive success from AI. 

The Implementation Questions Being Asked:

  • Will my enterprise see real business value come from AI?
  • We want to implement more use cases, how should our organization approach scaling AI
  • How do we determine which use cases are the most appropriate for our business goals? 

The first step in moving forward should be to clearly define AI and the role it plays in the enterprise. AI is the overarching technology with different branches that can all play a part in the overall AI ecosystem. For example, machine learning (ML) is just one of the many technology options that an enterprise can choose to adopt to create AI solutions (and, in fact, is one of the most popular). Technology options can range from ML to natural language processing, to deep learning, to reinforcement learning, and many of these options are interrelated. 

In order to pick out the technology applications that will lead to success, organization leaders should take a step back and investigate. As Gualtieri explains, “AI is a software that can mimic or exceed human intelligence to identify patterns, make decisions, and/or formulate new knowledge.” So in order to properly execute AI going forward, going back to ask the right questions that correlate with the capabilities builds a good foundation to scale from. Asking the right questions generates the clarity needed to isolate the technologies and particular use cases that match your business needs. 

  • Where do we make decisions? 
  • Where do we make predictions? 
  • Where/what are we responsible for identifying?

KITCHEN.00_01_23_23.Still005

What About Data? 

It goes without saying that data is important as it represents the real-world physical events that impact business decisions and, as IoT technology advances, rich data repositories quickly develop amongst business processes. Useful for a range of use cases, this data is intrinsically embedded throughout the enterprise. 

An enterprise has lots of intelligence and that intelligence is embedded in all of the people who work there, embedded in all of the data that gets generated by the systems, and embedded in all of the business processes and customer engagement.”  - Mike Gualtieri, VP and Principal Analyst, Forrester

Since the intelligence pre-exists AI integration, it is up to business leaders to craft strategies for incorporating and utilizing the available data within the framework of AI applications. AI allows an enterprise to scale and harness all of the diverse intelligence, but to do so in an organized manner, you need an AI platform. An accessible platform is the springboard for successfully scaled AI. Accessibility ensures that both technical and non-technical teams are provided with the tools and knowledge they need to properly interact with the AI platform as well as extract and communicate valuable insights. Dataiku is a unique platform because it allows for systemization through work streams both vertically (across different teams) and horizontally (within teams). 

→ Check out "Why Enterprises Need DS, ML, and AI Platforms" 

When to Start Using AI Platforms 

We have established that accessing data is a prerequisite to AI success. That said, companies don’t have to have all of their data ducks in a row before taking a dive into AI. Data will naturally flow in as a business grows and an established plan for the management of data influx will mitigate value loss. 

With the proper platform, AI is easier to adopt than ever before, and the sooner it is integrated into business processes, the sooner businesses will be able to extract real value from the integrations. In order to extract the most value, business strategies must evolve and day-to-day processes must shift to support pervasive AI and data efforts. This means ultimately committing to a full transformation. Digital transformations involve embracing both cultural and behavioral changes for the entirety of business operations, so the sooner an organization can adjust to this data-driven mind-frame the better. 

According to Mike Gualtieri at Forrester, nearly 100% of enterprises will use AI by 2025! The future of AI in many ways is already here, and it is up to organizations to fully step into their full AI potential to see the return on investment quickly. 

You May Also Like

Talking AI Democratization With Dr. Anastassia Lauterbach

Read More

6 Top-of-Mind Topics About AI & Trust in 2024

Read More

3 Concrete Ways to Drive AI ROI

Read More

Alteryx to Dataiku: The Visual Flow

Read More