Contact Us
hero-analytics-and-ai-trends-to-look-out-for-in-2021.jpg
  • Blog

Analytics and AI: Trends to look out for in 2021

Published

01/04/2021

In 2020, as COVID-19 stress-tested every organization, actionable data-driven insights became increasingly important to build business resilience. This will continue in 2021, and executives will need to accelerate artificial intelligence (AI)-powered digital transformation to stay ahead of the curve.

To help you prepare for the year ahead, let's explore the top analytics and AI trends:

1. The rise of industry-led augmented intelligence

When you combine machine intelligence and human judgment, you can realize augmented intelligence and turn insights into action at speed. When applied within the context of a given industry - also known as industry-led augmented intelligence - the benefits are even greater. For example, with Genpact as its partner and AI as its neural wiring, Envision Virgin Racing is connecting people, processes, and industry-specific racing knowledge to build an instinctive racing team that can make critical decisions at lightning-fast speed. Boardrooms across the globe need to replicate this approach.

2. From reports to stories

Businesses will consume insights not as reports, but as data stories – purpose-built recommendations for individual business roles – in 2021. While a regional business leader might receive insights on financial health indicator estimates as a percentage of global estimates, a C-level executive might receive these insights in the context of competitors' global performance. Moreover, we envision a future in which executives can request these complex insights from their virtual assistants, which will be ready to respond in real time.

3. More trustworthy AI

Organizations are facing substantial penalties, defamation, and lawsuits due to poor or no control over data use. In response, chief data and analytics officers (CDAOs), in partnership with Chief Data Ethics Officers, will develop a trustworthy AI framework. This framework rests upon five pillars – live model monitoring, compliant AI models, ethical AI models, explainable AI models, and change management. For example, chief data ethics officers must ensure that AI engines never make biased decisions based on sensitive information that would likely hit headlines if models went public.

4. Increased cloud adoption

The pandemic has shown us that cloud is the new normal – it's no longer just a nice-to-have – and every industry has explored a variety of cloud services. In 2021, it's time to double down. CDAOs and chief technology officers will identify and prioritize targeted use cases – backed by cloud partners across multiyear engagements – to realize the variety of benefits the cloud has to offer as we enter an increasingly virtual world of work.

5. Curated data ecosystems

Data is a differentiator, and organizations will continue mining for insights from various data sources. To meet this demand, consolidated third-party data offering platforms – known as data marketplaces – will furnish datasets, but at a price. While organizations will take interest in procuring data from these platforms, CDAOs will play a pivotal role in devising an overarching data governance plan to make sure their organizations don't lose control over the costs involved.

6. Elevated data literacy

Alarmingly, a huge amount of enterprise data is never analyzed. The reason? Limited data literacy across organizations. To support the democratizing of data in 2021, data literacy initiatives will increase and focus on how to read, work with, analyze, and communicate data effectively in the context of an individual's role. Leaders will either design internal learning programs – such as Genpact's Genome initiative – or partner with a commercial learning platform for professional development.

7. Increasingly pragmatic and human-centered AI

Employees are increasingly skeptical of industry-agnostic AI-based analytics solutions. Instead, they'll seek pragmatic AI models, customized to solve targeted and industry-specific business problems. For example, a demand forecasting engine for the eCommerce industry needs a completely different set of features from those that predict demand in manufacturing. And employees know it.

Outside the organization, CDAOs will increasingly view the acceptance rate of AI models by customers as a measure of success. Organizations will push AI innovation teams will to collaborate with social science, behavioral science, and decision intelligence experts to assess the connection between AI and customer experience. They will drive toward a virtuous cycle in which AI models and customers co-learn and adapt.

8. Enhanced MLOps

Enterprises want to shift from testing to operationalizing AI, but many are still reeling from the disruption COVID-19 has caused established models. For instance, product and service recommendation engines trained on pre-COVID data are likely to deliver outdated insights due to changes in customer behavior. To overcome this challenge, AI solutions teams will make machine learning operations (MLOps) practices an integral part of the overall AI model lifecycle to continuously monitor and retain AI models.

9. Composable and reusable assets

As AI use cases rise, costs rise too. Mature organizations will control and optimize these costs through composable and reusable assets. For instance, a 'synthetic data generation' asset added into invoice processing can learn how to create millions of representative invoice documents within seconds. AI practitioners will increase an organization's library of such assets to scale AI and simultaneously improve the employee and customer experience.

Ultimately, 2020 was a difficult year for every enterprise. But 2021 presents an opportunity to embrace and accelerate digital transformation using analytics and AI. Businesses that respond to these trends now will be best placed for a successful year ahead.

Visit our augmented intelligence page