Top 5 Streaming Realtime Analytics Use Cases
Lets talk about the use cases instead of focusing on the vendors
by Gus Segura
They say that depending on the work flow and type of data there are four main kinds of analysis: Descriptive: summary of data, Diagnostic: performance metrics, Predictive: using data trends to predict future outcomes and Prescriptive: recommend a course of action.
Today we’re going to look at some of the top use cases that depend on those types of analytics for streaming real-time decision support.
- IOT — Internet of Things Devices
Collect, monitor, visualize, aggregate, apply machine learning — predict. The promise of scada and plant floor data systems are reality today. We are pushing data into the pipeline and performing complex analytics in the pipe and at the edge “Edge Analytics” .We’re using this data to reduce cost, cut scrap, achieve quality goals, predict maintenance failures and more.
2. Retail Recommendation Systems
Lol, I wonder what the folks at Netflix think of some of the first recommendation systems they built vs. today. Can anyone say collaborative filtering? To be fair, I use the “Whats good on netflix now” Google query.
Think of a recommender as a smart filter that knows some of your base preferences and attempts to recommend products, services and movies based your past behavior likes, dislikes, viewing history and more. These systems can operate on single data points [like viewing history] or multiple points [combining behavior from multiple device].
3. Customer experience metrics
There are several fancy KPI’s for tracking customers: Net Promoter Score, CSAT — Customer Satisfaction, Customer Effort Score, Churn Rate, Retention Rate and many more. These KPI’s allow retailers and service providers to adjust the content, channel, path to product, etc.
Interesting that with Covid — the data from the last two years has to be teased in many cases to fit the trends previously.
4. Operations Analytics : Supply Chain
“Examples of operations analytics include predicting when machinery requires maintenance, uncovering fraud, understanding a decline in online orders in an e-commerce application, and tracking inventory.” — Spark Cognition.
A key part of being a subscription service partner for Advantis Data is establishing standard environments and setting up monitoring and alerting for services like Kafka Confluent, MySQL, Oracle, Databricks and many more. For more information — Please contact Advantis Data Services.
5. Risk Management: Mitigation
The security processes used to achieve business objectives. This includes vulnerabilities, threat assessment, priorities and of course monitoring and alerting. How do you decide what projects are the most important? How do you measure impact to business when you miss a risk? How leadership needs to be involved and where they need to be hands off.
Moving risk system into a real-time data pipeline is crucial to being able to react to changes in business and strategy. This is where leadership need to be hands-on in making good choices for the future.
Cloud Architect Insider ( Former Google, Amazon, HPE ) writes about products, strategies, tips to help you make smart choices about your cloud providers and partners. I work primarily with Advantis Data Services now. However, My reporting, recommendations and opinions are mine are reasonably objective. Face It. I’m a Data Scientist with a focus on language analytics. There always some bias.. but I do try :) Peace.
If anything from this article resonates or you would like more information please contact : Advantis Data for more information.
