CONNECTIONS / Databricks
Databricks + Datahash: activate the lakehouse you already report from.
Your Databricks tables hold purchases, leads, LTV scores, and offline sales your ad platforms never see. Signals reads them on a schedule, scoped through Unity Catalog, and delivers them hashed and consent-aware.
- ● databricks :: signals live
- > delta table sync main.sales.pos_ledger rows=48,900
- > unity catalog grant read_only
- hash sha256(email,phone) role=read_only
- route meta · google · tiktok · snap · linkedin
- ✓ 5 destinations synced · match 88%
The signal Databricks yields, and receives.
Every kind of signal that moves between Databricks and the platforms, through one hashed, deduplicated route.
- Offline conversions
- Transaction and in-store tables leave the lakehouse hashed and matched back to the campaign that sourced the sale.
- Lead conversions
- Pipeline-stage tables governed by Unity Catalog arrive as conversion events without a nightly extract.
- Custom audiences
- An LTV tier modeled once in a notebook becomes a hashed, scheduled audience, no CSV hand-off.
- Lead generation inbound
- Subscription and renewal events sync as the model rebuilds, so recurring revenue reaches bidding quickly.
Pick your platform. Every use case for it, in one place.
Every way Databricks data moves, grouped by platform and ranked by how many use cases Databricks supports.
Google 5 
Meta 4 
Snap 4 
TikTok 4 
LinkedIn 3 
OpenAI 1 - Offline conversions Google Offline Conversions API Closed-won revenue and purchase records, matched back to campaigns as offline conversions.
- Offline conversions Google Store Sales Closed-won revenue and purchase records, matched back to campaigns as offline conversions.
- Lead conversions Google Enhanced Conversions for Leads Lead and deal stage changes delivered as conversion events for lead-quality optimization.
- Custom audiences Google Customer Match Contact segments refreshed as hashed, scheduled audiences.
- Lead generation Google Forms Campaign lead submissions created as records in real time, deduped before creation.
- Offline conversions Meta Offline Conversions API Closed-won revenue and purchase records, matched back to campaigns as offline conversions.
- Lead conversions Meta Conversions API for CRM Lead and deal stage changes delivered as conversion events for lead-quality optimization.
- Custom audiences Meta Custom Audience Contact segments refreshed as hashed, scheduled audiences.
- Lead generation Meta Lead Ads Campaign lead submissions created as records in real time, deduped before creation.
- Offline conversions Snapchat Offline Conversions API Closed-won revenue and purchase records, matched back to campaigns as offline conversions.
- Lead conversions Snapchat Leads Conversions API Lead and deal stage changes delivered as conversion events for lead-quality optimization.
- Custom audiences Snapchat Custom Audience Contact segments refreshed as hashed, scheduled audiences.
- Lead generation Snapchat Lead Generation Ads Campaign lead submissions created as records in real time, deduped before creation.
- Offline conversions TikTok Offline Events API Closed-won revenue and purchase records, matched back to campaigns as offline conversions.
- Lead conversions TikTok CRM Events API Lead and deal stage changes delivered as conversion events for lead-quality optimization.
- Custom audiences TikTok Custom Audience Contact segments refreshed as hashed, scheduled audiences.
- Lead generation TikTok Lead Generation Campaign lead submissions created as records in real time, deduped before creation.
- Offline conversions LinkedIn Offline Conversions API Closed-won revenue and purchase records, matched back to campaigns as offline conversions.
- Lead conversions LinkedIn Leads Conversions API Lead and deal stage changes delivered as conversion events for lead-quality optimization.
- Lead generation LinkedIn Lead Generation Forms Campaign lead submissions created as records in real time, deduped before creation.
From authorization to delivered signal.
-
Connect
Connect with a scoped read-only role via Unity Catalog.
-
Map
Map columns to event fields and match keys in the visual mapper.
-
Hash and consent
Hashing and consent flags applied at read.
-
Deliver
Scheduled or incremental delivery, with per-destination match feedback.
Databricks questions, answered.
Does Datahash copy our lakehouse?
No. Reads are query-scoped through Unity Catalog and delivery is event-level, so nothing from your lakehouse is copied or warehoused on our side. Identifiers are SHA-256 hashed before anything leaves the Delta table, and a self-hosted Core deployment is available when data cannot leave your own environment, with the hashing done inside your perimeter.
How fresh is the sync?
On the cadence your notebook or job already keeps, from every 15 minutes to weekly, with incremental runs that read only the rows the model rebuilt. A table refreshed hourly delivers hourly and one rebuilt nightly delivers nightly, so job-cluster time stays low because no run re-scans the whole Delta table.
Who writes the SQL?
You do not have to write anything new. Point Signals at a Delta table or a view and the visual mapper aligns its columns to each platform's event and identifier fields, so an existing model becomes a destination without extra notebooks. Custom views are welcome when you want to shape the data before it leaves Databricks.
NEXT STEP
See your Databricks data working in every ad platform.
Ready to enable a use case, or still mapping what Databricks data could do? Our team helps you find the right place to start.
