The short answer is: you can’t. By definition, big data and excel are opposites. Big data is defined as “(…) data sets with sizes beyond the ability of commonly used software tools (..)” (i.e., Excel.)
Excel can support up to 1M rows (1,048,576). A small big dataset consists of billions of records.
You can use Excel as a query interface to big data (NoSQL) databases like Cassandra, but the data is not stored in your computer (your hard drive space is not enough) and Excel won’t be able to process it (is the database engine Excel is connecting to.)
Big data in Excel doesn’t mean you need a huge dataset to make your analysis somewhat more accurate. First, you need a solution like AWS Big Data platform, collecting data that’s important to you: customer transactions, server logs, applications events.
Then you need to pre-aggregate the data, so it fits the 1M rows limit Excel supports. You can aggregate by customers, and analyze one a time. Or aggregate by your application’s features.
In other words, you need to convert your big data (lowest granularity level) to small data (pre-aggregated level.) Then you can play with it as much as you want in Excel.