Alex Marquardt discusses the importance of calculating ingest lag and storing ingest time in Elasticsearch to enhance data observability and reliability. By storing an ingest timestamp for each document and calculating the difference between the event's occurrence and its arrival at Elasticsearch, users can identify delays and potential issues with remote system clocks or the ingest process itself. The blog outlines how to implement an ingest pipeline using Elasticsearch's set and script processors to add an ingest timestamp and calculate lag, which can be used in visualizations, monitoring, and alerting. Marquardt emphasizes the benefits of using the ingest timestamp over remote-generated timestamps, such as ensuring recent events are considered in visualizations and alerts, and avoiding issues from incorrect or malicious timestamp settings on remote systems. The approach enhances data accuracy and ensures that events are not missed due to ingest delays, offering a more reliable basis for monitoring and analysis.