Google Maps CSV exporter: managing exports across a year

How should you manage repeated Google Maps exports?

Lead Finder writes a new file each run, and the discipline around those files determines whether a year of extraction produces an asset or a folder nobody can use. Three practices carry most of the benefit. Name files so the territory, category, search term and date are all recoverable from the filename alone, because six months later nothing else will identify them. Keep every export rather than overwriting, since the value of the archive is the comparison between two dates and overwriting destroys exactly that. And split by territory at export time rather than producing one enormous file, because splitting later requires the geographic information to still be attached and interpretable.

Naming that still makes sense next year

A filename should carry the category, the territory, the search term used and the date. Anything less and a folder of exports becomes unusable within months, because the same category swept with different terms produces different lists and nothing inside the file records which term produced it.

That last detail is the one most often lost. Search term determines which slice of a category an export contains, so two files that look interchangeable may cover quite different populations, and only the filename preserves the distinction.

Splitting large territories

A single file covering an entire region is awkward to work and slow to open, but the reason to split at export time rather than afterwards is different: the split should follow how the list will be worked, which is usually geographic, and that decision is easiest while the sweep is still structured by area.

Splitting also makes partial refreshes possible. Re-sweeping one district rather than a whole region is cheaper in runtime and produces a comparison that is meaningful because the boundaries match the previous pull exactly.

Comparing two exports

The comparison is done on place ID, which is stable across exports where names and addresses are not. Rows present in the new file and absent from the old are openings; rows in the old and missing from the new have closed, moved or changed category.

That second group deserves checking rather than deleting, since a missing row sometimes means the business changed its primary category and is still trading. Treating absence as closure without verification removes live prospects from the list.

Related reading: getting the CSV format right , the quarterly refresh rhythm .