February 2019 Summaries

The payments market is rapidly evolving with increased digital transactions and lower costs, but this growth also leads to higher fraud rates. Companies are looking for ways to minimize fraud while maintaining seamless buyer experiences. Early methods like "decision engines" had limitations due to outdated systems and slow analysis times. Predictive analytics has emerged as a solution, but the challenge lies in selecting appropriate AI models and ensuring real-time data analysis at a global scale. Aerospike's "Hungry AI" offers an efficient solution by processing hundreds of terabytes and billions of events per day, quickly analyzing thousands of data points to either validate transactions or apply extra scrutiny.

Aerospike experienced a successful year in 2018 with significant growth and innovation. The company released multiple milestone products, including the industry's first non-relational database with strong consistency and high performance - Aerospike Enterprise Edition 4.0. This allowed them to compete in a broader range of enterprise-class applications while maintaining their lead over legacy NoSQL vendors. Their focus on edge-based applications has led to high customer retention rates, as well as positive reviews from Gartner Peer Insights. The upcoming Aerospike Summit '19 will showcase the company's continued growth and innovation within the enterprise market.

Digital transformation is challenging, with only a third of major IT implementations considered successful and nearly one-fifth deemed failures. Large organizations often struggle to balance business expectations and deployment reality due to their extensive portfolios of applications. John Dillon suggests a strategic approach based on the Pareto Principle, focusing on deriving knowledge from 20% of data and moving it to the edge for quick value capture. Instead of attempting massive projects, IT teams should build simple systems adjacent to legacy ones, pushing knowledge to the edge for high success probability and learning opportunities.

Retailers are increasingly adopting machine learning technology to improve customer experience, operational efficiency and reduce fraud. The growth in customer data and advancements in AI have positioned retail as a sector well-suited to benefit from these technologies. Retailers are looking to move beyond traditional segment-based rules analysis towards algorithmic decision-making that enables hyper-personalization at scale. Machine learning can be used to increase customer engagement, optimize in-store operations and reduce fraud by improving transaction security and identifying counterfeit products.