$ man deepline

Deepline

The orchestration spine for GTM plays. Durable datasets, waterfall enrichment, joins across providers like Apollo and Prospeo, and custom plays written as scripts. The layer where individual API calls become repeatable pipelines.

by Shawn Tenam

raw API access gets you data. it doesn't get you a pipeline. the hard part of programmable GTM is the connective tissue: what happens when provider A misses, how results from three providers join on one contact, where the dataset lives between runs so you don't re-pay for enrichment you already did. Deepline is that tissue. plays are durable, so a run that fails at record 400 of 1,000 picks up where it stopped instead of starting over. providers are interchangeable, so when one degrades you swap the waterfall order without rewriting the play. and because plays are scripts, they compose: the output of the enrichment play is the input of the qualification play.

Deepline sits under every multi-provider play I run. a typical play: pull a segment from the data lake, run Apollo-first email enrichment, fall back to Prospeo on the gaps, join the results back on domain, verify, and export to the sending layer. I bring my own provider keys, so the economics stay API-priced. custom plays handle the weird stuff: joins across LinkedIn scrape data and CRM records, projections that strip a 40-column dataset down to the 6 fields the campaign needs, fallback logic per provider. the spine means I write the play once and run it against any list.