PostgreSQL(Full Text Search) vs ElasticSearch

If PostgreSQL is already in your stack the best option for you is using the PostgreSQL full-text search.

Why full-text search (FTS) in PostgreSQL ?

Because otherwise you have to feed database content to external search engines.

External search engines (e.g. elasticsearch) are fast BUT:

• They can't index all documents - could be totally virtual
• They don't have access to attributes - no complex queries
• They have to be maintained — headache for DBA
• Sometimes they need to be certified
• They don't provide instant search (need time to download new data and reindex)
• They don't provide consistency — search results can be already deleted from database

If you want to read more about FTS in PostgreSQL there's a great presentation by Oleg Bartunov (I extracted the list above from here): "Do you need a Full-Text Search in PostgreSQL ?"

This as a short example how you can create a "Document" (read the text search documentation) from more than one table in SQL:

SELECT to_tsvector(posts.summary || ' ' || brands.name) 
FROM posts
INNER JOIN brands ON (brand_id = brands.id);

If you are using Django for your e-commerce website you can also read this article I wrote on "Full-Text Search in Django with PostgreSQL"