May 16, 2022
Table of contents
Technology Research Analyst
One of the greatest, albeit fictional, data analysts of all time Sherlock Holmes used to say that "It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts."
Indeed, a major achievement of humanity throughout its history has been the gradual shift from interpreting reality based on beliefs and conjectures to analyzing events or scenarios according to actual data. Introducing increasingly advanced methods and technologies has certainly fostered this approach. On the other hand, our world has also become way more complicated over the centuries, and peering into the full spectrum of underlying trends and phenomena driving its dynamics has proved a challenging task, even for the most insightful and well-equipped researchers.
To address this and many other issues related to the never-ending pursuit of truth, the data analytics discipline has undergone an enhancement process whose result is now known as "augmented analytics".
According to Gartner, which coined this term in 2017, augmented analytics involves the implementation of artificial intelligence and related technologies into the typical data analytics workflow to facilitate data collection and preparation, insight generation, and data query. This can be achieved by automating several phases of the data science and data analytics life cycle (including data management, data model development, and data visualization) and improving data analytics and business intelligence software with more advanced features to unlock the aforementioned upgrades.
Gartner has also contributed significantly to better defining the concepts of augmented analytics and augmented business intelligence in recent years, while keeping an eye on the core trends of these disciplines and their growing importance from a business perspective.
Tapping into Gartner's research and other relevant sources, let's explore the reasons behind the success of augmented analytics and augmented BI, their most important adoption areas, and the major platforms available on the market, along with a few tips for implementing them properly into your business.
Today, data is fuel for companies and any tool to capitalize on it represents an essential component in the arsenal of every business decision-maker. Unfortunately, traditional approaches to data analysis have failed to keep up with the times, as new potential threats are looming on the horizon:
As mentioned above, a suitable solution to address such problems was found in the growing integration of data analysis techniques and tools with artificial intelligence, machine learning, natural language processing (NLP), and other cognitive technologies. These advanced tech enablers allowed for a greater focus on automation, data democratization, and business-centered insight generation, driving the transformation process that turned standard analytical functions into so-called augmented analytics and augmented BI.
Clear signs of this transition can be spotted in the most recent augmented analytics market dynamics. According to Verified Market Research’s aforementioned study, for example, the global augmented data analytics and augmented BI market was worth $7.7 billion in 2020 and is set to reach $62.5 billion by 2028 with a CAGR of 29.8%.
This positive trend can be explained by the growing share of corporations relying on advanced analytical software solutions and business intelligence services to fully leverage their data assets, along with the need to democratize access to business insights and further improve productivity based on such information.
We have already detailed some of the key challenges that data scientists, data analysts, and corporate decision-makers face every day when it comes to turning data into business recommendations. We have also stated that augmented analytics, powered by artificial intelligence and other related technologies, can solve or at least significantly mitigate such issues. Now, we just need to explain how and what benefits it brings.
Here are three use cases representing the mainstays of augmented analytics and augmented business intelligence in a typical business scenario.
We've already pointed out that well before we analyze data, we need to find and collect it. This process, given the growing size and complexity of datasets in the age of digitization, requires the precision and patience of an amanuensis monk. Data scientists do their best, but unlike monks, they would rather not stick to the ora et labora (pray and work) rule by spending their entire lives performing mechanical tasks such as cleaning and labeling data.
Fortunately, much of these data preparation procedures can be automated through augmented analytics. Specifically, algorithm-based systems will take care of integrating multiple data sources (such as corporate databases or online advertising platforms) into a coherent data stream, transforming it into a unified storage format, loading it into data warehouses, and re-engineering it for future processing. This ensures way higher data management speed and accuracy compared to manual methods.
As the motion of sea currents reflects the action of multiple forces, such as the winds and the thermohaline circulation, the vast ocean of data is crossed by countless and rather mysterious dynamics that even the most attentive observers may struggle to grasp and decipher. In less poetic and more business-related terms, how can we identify key parameters among huge datasets, relate them to specific market trends, customer behavioral patterns, or other events, and therefore get practical suggestions to guide our decisions?
Considering thousands of metrics and identifying any relationships among them sounds extremely challenging and time-consuming for humans. Machine learning algorithms, on the other hand, can easily sift through data, spot patterns, anomalies, and correlations among variables, and build data models representing them to help us get a full picture of the phenomena we need to analyze. Furthermore, such algorithms fine-tune their models as they process new data, learning new relationships between data points through experience (as the term "machine learning" suggests).
As for actual deployment on the ground, for example, this means that an augmented analytics platform can monitor data related to your sales performance, ad reach on social media, and market stats. Based on such information, it may find out that a recent growth in sales didn't depend that much on your marketing initiatives but resulted from a favorable trend in your target market, therefore suggesting that you should rethink your advertising campaign despite your enthusiasm.
When you ask analysts for business suggestions, they will answer you in perfectly understandable human language. But the machines we rely on to automate the aforementioned insight generation processes can't do the same because, well, computers don't talk. Or maybe yes? Actually, this is a rhetorical question, as we all know that recent developments in artificial intelligence and NLP have enabled virtually any software or device surrounding us to understand and mimic human communication.
Augmented analytics platforms are no exception, as they provide users with a variety of functionalities to query data and receive clear business insights and recommendations in natural language (be it written or spoken) rather than code. NLP-based conversational interactions are typically complemented with data visualization features (such as dashboards, reports, and graphs) to highlight relevant metrics or trends and further simplify the interpretation of the platform's findings.
The most remarkable effect of this combination is the possibility of democratizing the way we approach and leverage data, ensuring access to valuable information for non-technical users.
Augmented analytics and augmented business intelligence represent the hottest trends in the analytics platform market, and all major players have implemented a consistent range of features in their products to enhance or automate the analytical workflow. This means that any organization willing to rely on such capabilities and therefore step up data-driven decision-making can count on the full gamut of software solutions to choose from.
Despite a rather competitive landscape that seems destined for further fragmentation in the coming years, we may identify 5 key players in terms of market share, as pointed out by Mordor Intelligence in its 2021 Augmented Analytics Market report.
Selecting a platform that fits your business needs is not a simple task. To help you choose the right one, it's worth defining the core functionalities that you should look for in an augmented analytics and augmented BI solution and examining the strengths and weaknesses of the major tools on the market.
As for the first point, Gartner's analysts compiled a list of recommended augmented analytics features, which include:
Regarding the second aspect, namely the pros and cons of the five major platforms for augmented analytics and augmented business intelligence, here's a brief roundup.
Microsoft Power BI
At present, Microsoft's flagship platform can easily be recognized as the market leader in augmented analytics. Power BI is fairly priced, offers solid automated ML functionalities, and can work in synergy with other Microsoft products, such as Office 365 and Teams.
However, it's still far from being the "One tool to rule them all" due to a general lack of flexibility in terms of deployment options. Power BI runs only in Azure and its on-premises functional capabilities are pretty limited.
Organizations deploying Qlik's lead product for augmented analytics can count on its powerful Associative and Cognitive Engines to get context-aware insights in self-service mode. Furthermore, Qlik Sense offers both cloud and on-premises deployment, flexible data visualization and storytelling capabilities, and a constantly expanded range of features thanks to new acquisitions.
On the other hand, its product licensing model is rather complex and the aforementioned acquisitions have resulted in overlapping or poorly integrated functionalities.
Previously known as Einstein Analytics, Tableau is a business-centered augmented analytics platform providing decision-makers with advanced visualization functionalities to help them explore data and share valuable insights. Another core strength of Tableau for data science is its vibrant user community of customers and partners sharing their expertise.
Anyway, all this comes at a (relatively high) cost, which is Tableau's premium pricing. Furthermore, based on Gartner Peer Insights' feedback, its overall service and support quality is slightly lower than the competition, especially in terms of response speed.
IBM Cognos Analytics with Watson
IBM's watchword is "flexibility", since its augmented analytics solution offers a variety of on-premises, cloud and hybrid deployment models, supports several IaaS platforms (such as Azure, GCP, AWS, and IBM Cloud), and covers a comprehensive range of use cases. These include automated insight generation, NLP-based data storytelling, visual data exploration, and other augmented analytics features.
The flip side of the coin is that Cognos' pricing per user is higher than other cloud alternatives and that IBM lacks key enterprise-focused applications in its product offering, including digital workplace software solutions.
SAP Analytics Cloud
SAP's answer to the growing demand for augmented analytics tools from business decision-makers is a cloud-based platform that boasts excellent usability and an intuitive user interface. It also combines advanced NLP-powered data query and automated insight generation features with a one-of-a-kind planning functionality based on predictive analytics.
Unfortunately, SAP's augmented analytics solution is essentially a cloud-native platform that requires additional tools from the same ecosystem (but with completely different support and maintenance schedules) in order to be deployed on-premises.
Let's close our overview with another bunch of noteworthy contenders that might carve out a prominent place in the augmented analytics and augmented business intelligence market in the future.
|Google Loocker||Direct data query of cloud databases||Requires coding and more advanced expertise|
|Domo||Great connectivity with corporate applications and rapid deployment||Lack of a comprehensive app ecosystem|
|ThoughtSpot Everywhere||Strong focus on NLP and automated insights||Limited global reach and broader ecosystem|
|Sisense||Extensible architecture and NLP-based interactivity||Lack of third-party resources and support|
|Oracle Analytics Cloud||End-to-end solution for data management and analytics||Embedded analytics features work only with Oracle apps|
Augmented analytics is revolutionizing the way we explore data and use it as a potential asset. However, it's not a panacea, and the general recommendations for facilitating the implementation of data analytics into your corporate workflow still hold true for its more advanced counterpart:
Such guidelines should help you unleash augmented analytics’ full potential, streamline the workflow of your analysts and managers, and make the most out of your data assets. In this regard, let me conclude our overview with another famous quote from good old Sherlock.
“Data! Data! Data! I can’t make bricks without clay.”
Well, he was certainly right. But with better instruments, shaping bricks will be significantly easier. And nowadays, the most powerful tool at our disposal is augmented analytics.
Take a look at our overview of the top use cases, benefits and implementation scenarios of real-time big data analytics.
Itransition’s BI team presents the reactive data analytics strategy for uncertain times. Learn more.
Itransition developed a data analytics system for decision-makers to utilize information collected from heterogeneous web sources.
Learn how Itransition helped the leader in maritime data analytics to migrate their big data and extended the system with analytical features.
We look at the current approaches to automated data collection and their effectiveness for text, audio and video information extraction.