Analytics is a systemic way of discovering data for aligned or non-aligned patterns and presenting it in a visualized manner that the human mind cannot miss the apparent trends that presentation is highlighting. The summarised highlights of data presented in charts and graphs pave the way for easy understanding. Analytics became popular right since accumulatio of business data coupled with the introduction of specialist analytics tools in the marketplace that claimed exclusive data discovery familiarity. As these tools became popular, we noticed corporate houses performing various analytics for top management by way of relevant charts. This kindled human reaction to those visualization and thus was born data-driven decision making. Senior management makes strategic decisions based on data visualization presented to them. However, in the early days of these tools, the analytics visualization was restricted to senior management, and scarcely, the operational team had access to these analytics platforms. Dashboards were not used by the operational team, least to create it by themselves for their own data analytic needs. However, all the analytics were batch driven then, not providing helpful insights on a real-time basis. It is like the report card of our children which we get every quarter from the school to know what scores they have got and being blank until the end of the quarter to see how the child is progressing in studies. With the advent of the Millennial and Gen Z workforce, the appetite for self-service and drive for making decisions on their own gained momentum. Their independent cum collaborative working styles required that they own their data, network about their thoughts on data analytics in close circles, and pursue actions from their consideration. Self-service analytics and on-demand analytics came into play to their advantage. On-demand analytics is not possible in batch mode, and the absence of embedded analytics was an apparent frustration to them. All the frontline analytics vendors rushed in to fill the gap with connectors that can export data almost in real-time to provide real-time analytics. Some created common platforms to connect reputed ERP, and other popular cloud software and others provided a generic model of configuring connectivity between systems to pull data almost instantaneously. The Embedded Analytics became the basic norm, not even as a unique feature beyond the primary offering. However, still, the third-party embedded analytics had its limitations of not being fit for fast throughput online customer-facing data roll situations like stock market transactions or online revenue management systems or the like. Further, some of the online ERP systems cater to high volume eCommerce transactions, and embedded analytics fell short of configuring for alerts integrated with workflows linked to those transactions. It is in this context, natively embedded analytics came in, developed natively by software product owners, not to miss even the split of seconds that is involved in the transfer of data to the third party data model and produce analytics based on such transferred data. Natively embedded analytics uses the same database rather than using another database replicated online from the central action database. Natively embedded analytics stands superior to third party embedded analytics. However, third party specialist embedded analytics were much superior in product features and performance being specialist products. The trade-off then was what to choose between embedded and natively embedded or hybrid of them. The natively embedded has the advantage of showcasing what-if situations as transactions are captured even before a transaction commit is made if situation demanding and warranting. While one may not do transaction-based analytics in an online manner, there may be situations that require predictive analytics for transactions in the offing. A natively embedded one has an edge to provide that instantly like an instant coffee. While building Retail ViVA, our Retail Management ERP software, we developed a natively embedded analytics platform of our own so that users need not miss out on their abilities to use the analytics even for transaction-based decisions making. While the senior management may be involved in data-driven decision making, the operational team may be involved in transaction-based decision making. Further, a unified database used for analytics produces one truth which is core in corporate decision-making. While embedded analytics war is on between various specialist business intelligence vendors, large ERP product vendors have created their own native embedded analytics given largeness and complexities of their data model. Who will win this war is an interesting question and time alone will answer? Whoever be it, the native has an edge, being really at the Edge. Maybe, a hybrid model may best fit to serve multi-purpose needs of modern growing retail enterprise. Written by (Ragu)nathan Kannan https://retailviva.com/
