DATA CLOUD
One customer, one profile, across every system
Ingest, resolve, harmonise, activate. We start from the decision the data has to change, then build back to the sources.
Data Cloud projects go wrong when they start as data projects. Everything gets ingested, a lake forms, and eighteen months later nobody can name a decision that improved.
We start at the other end: name the decision — which customers to call, which case to prioritise, which offer to suppress — then work backwards to the minimum set of sources, the identity resolution ruleset, and the calculated insights that decision needs. Activation into the org, into Marketing Cloud, or into an agent’s context is part of the build, not a later phase.
What you get
- Scoped by decision, so value arrives in weeks rather than after a full ingestion programme.
- Identity resolution rules you can inspect, tune and explain.
- Calculated insights that land where someone will see them — a record page, a journey, an agent’s context.
- Consent and retention modelled as first-class, not retrofitted.
- Zero-copy patterns with Snowflake or BigQuery where they save duplication and cost.
- Segment and profile quality monitored, with match-rate reporting.
Features
Data stream ingestion (CRM, web, mobile, ERP, support, files) · Data model mapping to the Customer 360 model · Identity resolution rulesets and match testing · Calculated and streaming insights · Segmentation and activation targets · Zero-copy federation (Snowflake, BigQuery, Databricks) · Consent and retention policy modelling · Data Cloud-triggered flows and Agentforce grounding
How we run it
01 Decision definition and value case · 02 Source inventory and profiling · 03 Model mapping and ingestion · 04 Identity resolution design and match testing · 05 Insight and segment build · 06 Activation and measurement
What you receive
Decision-to-data map · Source inventory with quality scores · Mapping specification · Identity resolution ruleset and match-rate report · Insight definitions · Activation configuration · Consent and retention register · Monitoring dashboard
Common questions
Is Data Cloud a data warehouse?
No. It’s a customer-profile layer. It complements a warehouse and, with zero-copy, often avoids duplicating it.
How long before it’s useful?
For a single well-defined decision with two or three sources, six to ten weeks. Broad “ingest everything” programmes take far longer and deliver less.
How is identity resolution tuned?
With a labelled test set of known duplicates and known distinct pairs, so match and false-match rates are measured, not assumed.
Do we need it for Agentforce?
Not strictly, but grounding quality is the main driver of agent quality, and a resolved profile is the most reliable grounding you can give it.
Start here
Tell us what your CRM is getting wrong
Thirty minutes with a consultant, not a salesperson. You’ll leave the call with at least one thing you can fix yourself — whether or not you hire us.
Prefer email? hello@zeptglobal.com · Typical reply time: one business day