This piece first appeared in CIO.
Business analytics continues to be a hot segment in the enterprise software market and a core component of digital transformation for every organization. But there are many specific advances that are at differing points along the continuum of market readiness for actual use.
It is critical that technology leaders recognize the difference between mature trends that can be applied to real-world business scenarios today versus those that are still taking shape but make for awe-inspiring vendor demos. These trends fall into categories ranked from least to most mature in the market: artificial intelligence (AI), natural language processing (NLP), and embedded analytics.
Artificial augments actual human intelligence
The hype and excitement surrounding AI, which encompasses machine learning (ML) and deep learning, has surpassed that of big data in today’s market. The notion of AI completely replacing and automating manual analytical tasks done by humans today is far from application to most real-world use cases. In fact, full automation of analytical workflows should not even be considered the final goal — now or in the future.
The term assistive intelligence is a more appropriate phrase for the AI acronym, and is far more palatable for analysts who view automation as a threat. This concept of assistive intelligence, where analyst or business user skills are augmented by embedded advanced analytic capabilities and machine learning algorithms, is being adopted by a growing number of organizations in the market today. The utility of these types of smart capabilities has proven useful in assisting with data preparation and integration, as well as analytical processes such as the detection of patterns, correlations, outliers and anomalies in data.
Natural interactions improve accessibility of analytics
Natural Language Processing (NLP) and Natural Language Generation (NLG) are often used interchangeably but serve completely different purposes. While both enable natural interactions with analytics platforms, NLP can be thought of as the question-asking part of the equation, whereas NLG is used to render findings and insights in natural language to the user.
Of the two, NLP is more recognizable in the mainstream market as natural language interfaces increasingly become more commonplace in our personal lives through Siri, Cortana, Alexa, Google Home, etc. Analytics vendors are adding NLP functionality into their product offerings to capitalize on this consumer trend and reach a broader range of business users who may find a natural language interface less intimidating than traditional means of analysis. It is inevitable that NLP will become a widely used core component of an analytics platform but it is not currently being utilized across a broad enough range of users or use cases to be considered mainstream in today’s market.
On the other hand, NLG has been in the market for several years but only recently has it been incorporated into mainstream analytics tools to augment the visual representation of data. Many text-based summaries of sporting events, player statistics, mutual fund performance, etc., are created automatically using NLG technology. Increasingly, NLG capabilities are also being used as the delivery mechanism to make AI-based output more consumable to mainstream users.
Recently, analytics vendors have been forging partnerships with NLG vendors to leverage their expertise in adding another dimension to data visualization, where key insights are automatically identified and expressed in a natural language narrative to accompany the visualization. While the combination of business analytics and NLG is relatively new, it is gaining awareness and traction in the market and has opened the door to new uses cases for organizations to explore.
Embedded analytics brings insights closer to action
The true value of analytics is realized when insights can inform decision-making to improve business outcomes. By embedding analytics into applications and systems, where decision-makers conduct normal business, a barrier to adoption is removed and insights are delivered directly to the person who can take immediate action.
Modern analytics platform vendors have made it incredibly easy for organizations to adopt an embedded strategy to proliferate analytic content to line-of-business users previously unreachable by traditional means. And organizations are now extending similar capabilities to customers, partners, suppliers, etc., in an effort to increase competitive differentiation and, in some cases, new revenue streams through monetization of data assets and analytic applications.
These innovations present technology leaders with a unique opportunity to lead their organizations into an era where data analysis is the foundation for all business decisions. Every organization will embark on this journey at its own pace. Some will be early adopters of new innovations and some will only adopt when the majority of the market has successfully implemented.
Ultimately, organizational readiness to adopt any new technology will be determined by end users and their ability and willingness to adopt new innovations and embrace process change.