An ecommerce company wants to collect user clickstream data from its website for real-time analysis. The website has fluctuating traffic throughout the day. The company needs a scalable solution that adapts to varying traffic levels. Which solution meets these requirements?

Kinesis Data Streams in on-demand mode is ideal for fluctuating traffic because it automatically scales capacity without requiring shard management, making it cost-effective and operationally simple. Lambda is a serverless compute service that can process data in real-time, scaling automatically with the incoming data volume from Kinesis, perfectly matching the need for real-time analysis. Kinesis Data Firehose is primarily for delivering streaming data to destinations like S3, Redshift, or Splunk, not for real-time processing with Lambda directly from the stream. AWS Glue is an ETL service, better suited for batch processing or preparing data for analytics, not typically for real-time stream processing from Kinesis. Kinesis Video Streams is specifically designed for video data, not clickstream data. Amazon Managed Service for Apache Flink is a powerful tool for real-time analytics on streaming data, but it's a processing engine, not a data capture service like Kinesis Data Streams.