I believe one approach would be to digest part of the double strand to generate the single strand.

I assume that you are asking for a longer overhang than produced by a restriction enzyme.

There are several methods to do this, including USER reaction mix, nicking enzymes with specific recognition sites (not sure whether these methods leave 5' or 3' overhangs but it's easy to check).

You could also use a 3' exonuclease for a limited reaction time.

Put a restriction site at the beginning of the reverse primer that leaves an overhang when cut. PCR cleanup after amplification, then digest, then gel purify.

Design the primer as you normally would, but then add the restriction site and two or three spacer bases at the beginning. Some enzymes may need up to 6 bases added. See this page for spacer requirements:

https://www.neb.com/tools-and-resources/usage-guidelines/cleavage-close-to-the-end-of-dna-fragments

Sample primer:

Reverse primer: 5'-ATCGAGTTCAGGTGCACG-3'

With BamHI cut site: 5'-GTC GGATCC ATCGAGTTCAGGTGCACG-3'

I agree with Jason Rosenzweig, the easiest way to add any overhang, 5' or 3', to your PCR product is to add the overhang in the primers and then proceed with the PCR reaction as you normally would.

This is very useful for doing restriction enzyme based cloning of your PCR product, if you do not want to struggle with blunt-end ligations.

You need to be a bit more specific if you want a sensible answer.

What sort of overhang are you talking about? A single nucleotide? A compatible restriction site "sticky end", Or something a bit longer?

How long is your PCR product?

I fully agree with Matthew. Also, if you don't need an overhang with a discrete length, you can shorten the extension step in you last PCR cycle to get plenty PCR products with 5' overhang.