AI solutions will see greater success by reducing friction and helping solve defined business problems.
How we got here
We’re experiencing a golden age of data and technology—and there is no sign of it slowing. Artificial intelligence (AI) technology continues to improve: machine learning (ML) models are processing trillions of lines of data, natural language processing (NLP) advancements are moving towards understanding human intent, and algorithms are getting faster. We’re seeing more simple, repetitive tasks be automated, giving rise to new opportunities to enable humans to do what they do best: reasoning critically and understanding data in context.
As innovation accelerates, so do AI investments and adoption, with 99% of Fortune 1000 companies planning to invest in data and AI in the next 5 years. Business and IT leaders believe it’s critical to the future survival of their business. But many considerations factor into the long-term success and sustainability of AI solutions: increasing amounts of data, costs of maintaining this technology, difficulty in staffing highly-specialized roles, and scaling AI pilots to widespread adoption.
![](https://www.tableau.com/sites/default/files/2022-01/data-trends-2022-Quote1.png)
84% of C-suite executives believe they must leverage artificial intelligence to achieve their growth objectives, yet 76% report they struggle with how to scale.
Businesses recognize that they need to do more to innovate and better serve their customers. While AI opens up opportunities, most investments have yet to deliver on their potential value. In 2022, AI technologies will reach new levels of success through human augmentation: assisting and enhancing people to think critically and make data-driven decisions. Think of analytics and AI as being supporting members of the team.
Data Culture and data literacy—the ability to explore, understand, and communicate with data—also help organizations figure out their AI and ML strategy and perspective. These change management and workforce development efforts affect how they’ll stay competitive and manage the spectrum of human augmentation, beginning with questions like:
- What tasks will be completely automated with AI technology?
- Examples of automation that free up people to focus on more sophisticated tasks: Basic language translation and image editing. Rather than spending hours manually editing a photo to change the background, editing can be done with default image editing technology that incorporates AI to handle lighting and blending techniques. These automated tools facilitate new levels of creativity.
- Which tasks will be semi-automated and require human involvement and interpretation?
- Examples of AI that distills useful patterns and insights to empower people to make data-driven decisions in context:
- To more accurately weight climate and pandemic models, ML techniques are applied to help researchers understand trends, impacts, and patterns to help with policy decisions.
- Machines can inspect unlabeled voice data (e.g. customer calls) using NLP and ML algorithms to better understand user intent, adding relevant categories and labels. These signifiers and semantics inform people of what action to take next.
- Examples of AI that distills useful patterns and insights to empower people to make data-driven decisions in context:
![](https://www.tableau.com/sites/default/files/2022-01/data-trends-2022-Quote1.png)
Organizations that invest in change management were 60% more likely to report that AI initiatives exceeded expectations and 40% more likely to achieve outcomes than those that don’t.
Having common behaviors, beliefs, and data skills also facilitate the ability to scale AI solutions, supporting sustainable implementation and innovation. In a recent report, Gartner® found that the “lack of skills was cited as the No. 1 challenge to the adoption of artificial intelligence and machine learning.” Because investing in the development of your people and AI techniques is an ongoing process, constantly evolving alongside the technology. Having your entire workforce in agreement and appropriately skilled may mean the difference between seeing AI proofs of concept become scalable, practical applications or fail entirely.
![](https://www.tableau.com/sites/default/files/2022-01/data-trends-2022-Quote1.png)
Companies in our study that are strategically scaling AI, report nearly 3x the return from AI investments compared to companies pursuing siloed proof of concepts.
Where we're going
In collaboration with IT leadership, business leaders have an opportunity to drive data and AI strategies grounded in business context. For AI technology to be relevant, maintainable, and explainable, it needs to empower people and be tied to business strategy and goals. We’ll see AI solutions move from a proof of concept model to widespread implementation for business- and industry-specific use cases.
Various industries are developing and using AI in innovative ways. A recent study by KPMG examined AI deployment across five industries (retail, transportation, healthcare, finance, and technology), finding that for “91% of healthcare industry respondents, AI is increasing access to care for patients.” And although most businesses manage their supply chains manually, “those that adopt AI in the coming months and years will achieve significant competitive differentiation,” according to the Harvard Business Review.
Thanks to cloud computing, AI has become more affordable and accessible, leading to greater innovation across experiences and industries. And with an additional focus on business success, we’ll see solutions which combine different AI techniques to achieve better results (also known as composite AI) added to support people, specifically “tuning” this intelligence to specific workflows.
![](https://www.tableau.com/sites/default/files/2022-01/data-trends-2022-Quote1.png)
You must deliver creative new uses of technology to enable your organization to scale digitalization rapidly. You must collaborate with business and other IT leaders and create teams that fuse business and IT skills from various disciplines.
Workflows will be brought to life and made more efficient with shared skills, mindset, and values—Data Culture and data literacy—which facilitate the ability for people to complete new, more sophisticated data science and analytics tasks required for AI success.
Recommendations
1. Treat AI as a team sport. Identify what tasks and functions would best support human augmentation by saving people time or elevating their skills or expertise. Begin by looking at your customers’ needs and pain points to understand where your AI solution can add value for them. Ask yourself these questions to see if a proof of concept or pilot is worth developing:
- How many customers have similar needs or experience these same issues?
- How often are these issues happening?
- Are these issues solvable with AI technology?
2. Focus on business use cases and success factors to move from proof of concept and successfully scale.
- Drive intentional and contextual AI by connecting solutions to real business problems with defined goals to realize their value.
- Identify where AI can enable and reduce friction. Avoid trying to enable AI in all aspects of your product suite—you’ll struggle to scale by spreading your resources too thin.
- Be wary of “shiny,” pipe-dream projects. While attractive, they rarely move beyond proof of concept. And tune out the noise by setting realistic time and scope expectations for AI projects, balancing all resources like budget, time, highly-technical staff, and infrastructure.
3. Invest in data literacy to upskill and develop your workforce.
- Poor data quality results in inaccurate and ineffective AI solutions. And a data-literate workforce can improve issues with data quality, building and/or training AI, ML, NLP, etc. algorithms and models with accurate, timely, and relevant data.
- Even a basic, “Data 101” training, whether developed internally or offered through a third-party, can give business users what they need to answer their questions. This will reduce the number of simple or lower-stakes analytics requests that go to advanced analytics and data science teams—freeing them up to work on high-value, large-scale projects.
1 Gartner®, Maximize the Value of Your Data Science Efforts by Empowering Citizen Data Scientists, Pidsley, David and Idoine, Carlie, 7 December 2021. GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the US and internationally and is used herein with permission. All rights reserved.