Top 10 Data and Analytics Trends for 2021

Enterprise organizations have embraced the ideas behind advanced analytics technologies over the past several years, beginning with buzz words like big data and moving onto topics such as machine learning and artificial intelligence. But the promise of these technologies can sometimes get lost in the reality of implementing them in the real-world enterprise. Depending on what survey you are looking at, how you define the technologies, and what questions you ask, enterprise organizations' adoption of advanced analytics, machine learning, and AI varies quite a bit.

But the technologies have captured the attention of both the IT pros in the trenches and the top enterprise executives who recognize its promise for everything from cutting costs, to increasing revenue, to accelerating innovation and improving competitiveness in the market.

Last year Gartner said that enterprise AI deployments to production hit 19%. But as the technology becomes more mainstream through more support from vendors, a bigger talent pool, and a host of technology advances, enterprises will be in a better position to put artificial intelligence to work in a number of ways that hadn't been considered.

Read more on data and analytics here:

Data Science: How the Pandemic Has Affected 10 Popular Jobs

How to Create a Successful AI Program

The State of Chatbots: Pandemic Edition

Analytics Salaries Steady Amid COVID Crisis

With that in mind, during its recent Gartner IT Symposium, the analyst firm unveiled its Top 10 Strategic Technology Trends in Data and Analytics, 2020, a list designed to take organizations "from crisis to opportunity," as enterprises recover from the effects of the pandemic on business and IT initiatives.

Rita Sallam

Here are the trends presented by research VP Rita Sallam during the Gartner Symposium.

1. Smarter, faster, more responsible AI. Gartner is forecasting that 75% of enterprises will shift from piloting to operationalizing AI by the end of 2024, driving a 5x increase in streaming data and analytics infrastructures. There are challenges with current approaches. Pre-covid models based on large amounts of historical data may no longer be valid.

But the disruptions in AI will enable learning algorithms such as reinforcement learning, interpretable learning such as explainable AI, and efficient infrastructures such as edge computing and new kinds of chips.

2. Decline of the dashboard. Data stories, (not dashboards) will become the most widespread way of consuming analytics by 2025, and 75% of these stories will be automatically generated using augmented analytics techniques. AI and machine learning techniques are making their way into business intelligence platforms. In dashboards users have to do a lot of manual work to dive into further insights. But these data stories provide the insights without requiring the user to perform their own analysis.

3. Decision intelligence. More than 33% of large organizations will have analysts practicing decision intelligence including decision modeling by 2023. Gartner defines decision intelligence as a practical domain that includes a wide range of decision-making techniques. It encompasses applications like complex adaptive systems. It includes a framework that brings together traditional techniques like rules-based approaches together with advanced techniques like AI and machine learning. This enables non-technical users to alter decision logic without involving programmers.

4. X analytics. The "X" here is a stand in for any number of words that go before analytics. Gartner said that AI for video, audio, vibration, text, emotion, and other content analytics will trigger major innovations and transformations in 75% of Fortune 500 companies by 2025. X stands for the type of analytics such as video analytics or audio analytics. This will open new opportunities for analytics because this type data has not been fully leveraged by most organizations. But the effort to leverage it is growing. Sallam said that AI techniques and use of AI in the cloud are maturing to expand adoption and the impact of X analytics. There are a host of untapped use cases here, such as image and video analytics for supply chain optimization, or video and audio analytics for weather or for traffic management.

5. Augmented data management: metadata is "the new black." Organizations utilizing active metadata, machine learning, and data fabrics to dynamically connect, optimize, and automate data management processes will reduce time to data delivery by 30% by the year 2023.

AI techniques are being used to recommend the next best action, or auto-discovery of metadata, or auto-monitoring of governance controls, among many others. This is enabled by a concept Gartner is calling data fabric. Gartner defines data fabric as something that uses continuous analytics over existing, discoverable, and inferenced metadata assets to support the design, deployment, and utilization of integrated and reusable data objects, regardless of deployment platform or architectural approach.

6. Cloud is a given. Public cloud services will be essential for 90% of data and analytics innovation by 2022. Cloud-based AI will increase 5x between 2019 and 2023, making AI one of the top workload categories in the cloud. This trend started well before the pandemic, but COVID-19's impact on the enterprise has certainly accelerated it. Cloud vendors themselves are supporting data to insight to models within their portfolios. From a cloud vendor perspective, the more the cloud providers get you to do data and analytics in their cloud, the more you will also do compute in their cloud. On the enterprise side, using public cloud services makes it possible for organizations to do the work faster. Cloud is the new stack, according to Sallam.

7. Data and analytics worlds collide. Non-analytics applications will evolve to incorporate analytics over the next few years, according to Gartner. By 2023, 95% of Fortune 500 companies will have converged analytics governance into broader data and analytics governance initiatives. Even sooner, by 2022, 40% of machine learning model development and scoring will be done in products that do not have machine learning as their primary goal, Sallam said. Analytics and BI vendors are adding data management capabilities. Data management vendors are adding data prep. Expect to see more convergence in the near future.

8. Data marketplaces and exchanges. Gartner forecasts that 35% of large organizations will be either sellers or buyers of data via formal online data marketplaces by 2022. That's up from 25% in 2020, according to Sallam. This trend is all about accelerating cloud, data science and machine learning, and AI, she said.

9. Practical blockchain (for data and analytics). Gartner believes that within the data and analytics realm, blockchain will be used for vertically specific, business-driven initiatives such as smart contracts. It won't be used to replace existing data management technologies, according to Sallam. Blockchains are not inherently more secure than alternative data sources. Gartner forecasts that by 2023, organizations using blockchain smart contracts will increase overall data quality by 50%, but reduce data availability by 30%, conversely creating positive data and analytics ROI.

10. Relationships form the foundation of data and analytics value. Graph technologies will facilitate rapid contextualization for decision-making in 30% of organizations worldwide by 2023, according to Gartner. Graph databases and other technologies put the focus on relationships between data points. Those relationships are critical for most things we want to do with data and analytics, Sallam said. We want to know what are the drivers of this particular outcome? What thing did people buy after they bought an umbrella? What things do people buy at the same time?  But most relationships are lost when using traditional storage approaches. Joining relational tables together uses a lot of resources and degrades performance. Graph technology preserves these relationships and increase the context for machine learning and AI. They also improve the explainability of these technologies.