Definition
Dataflow Step is a Salesforce analytics concept that supports the creation of data visualizations and business intelligence outputs. It transforms CRM data into insights that help teams optimize their strategies and operations.
Real-World Example
At their company, a business intelligence manager at Apex Analytics leverages Dataflow Step to transform raw Salesforce data into actionable business intelligence. After setting up Dataflow Step, leadership has real-time visibility into pipeline health, team performance, and customer trends, enabling faster and more confident decision-making.
Why Dataflow Step Matters
A Dataflow Step is an individual operation within a CRM Analytics (formerly Tableau CRM or Einstein Analytics) dataflow that transforms raw Salesforce data into analytics-ready datasets. Each step performs a specific function such as extracting data from an object (sfdcDigest), filtering records (filter), computing new fields (computeExpression), combining datasets (augment or append), or flattening hierarchical data (flatten). Dataflows chain these steps together in sequence to build the curated datasets that power dashboards and lenses, making them the backbone of the analytics data pipeline.
As analytics requirements grow more complex, the design of individual dataflow steps becomes critical to both data accuracy and pipeline performance. Poorly constructed dataflows with unnecessary steps, unfiltered extracts that pull millions of irrelevant records, or computeExpression steps with expensive logic can cause dataflows to time out or consume excessive processing quotas. Organizations that invest in optimizing their dataflow steps by filtering early, minimizing the number of fields extracted, and parallelizing independent operations see faster dataset refreshes and more responsive dashboards. Conversely, neglecting dataflow optimization leads to stale analytics data, frustrated business users, and escalating processing costs.
How Organizations Use Dataflow Step
- Apex Analytics — Apex Analytics built a 12-step dataflow that extracts Opportunity, Account, and Product data using sfdcDigest steps, joins them with an augment step, computes win rate percentages with a computeExpression step, and filters out test records. The resulting dataset powers a pipeline health dashboard that refreshes every four hours, giving leadership up-to-the-day visibility.
- Meridian Retail — Meridian Retail uses a flatten step in their dataflow to transform hierarchical territory data into a flat structure that can be joined with sales records. Combined with a computeExpression step that calculates year-over-year growth, their regional managers see territory performance comparisons without needing the BI team to run custom queries.
- CloudServe SaaS — CloudServe SaaS optimized their analytics dataflow by adding filter steps immediately after each sfdcDigest extraction to exclude closed-lost opportunities older than two years. This single optimization reduced their dataflow execution time from 45 minutes to 12 minutes and kept their daily dataset refresh within the processing quota.