Skip to main content
Intempt

Maximize enrichment coverage without paying for every provider on every record

A single enrichment provider always misses some accounts. This falls through a cascade of providers, and finally AI research, only spending more when the cheaper source actually missed.

Sellworkflow-builderB2BAdvanced7 steps2 outputs
1
2
3
4
+3 more
Workflow Step
Run this recipe

What it does

Runs a multi-source enrichment cascade - try the primary provider, fall through to secondary, then tertiary, then an AI-research fallback for unstructured discovery - maximizing coverage while minimizing per-record cost.

You get

maximum enrichment coverage at minimum per-record cost

How it works

1

Build the Waterfall Enrichment Workflow

Create Workflow

Create a workflow 'Waterfall account enrichment' triggered by account_created OR scheduled refresh for stale accounts (no enrichment update in 90 days). Goal: maximize fill rate on key account attributes (firmographics, technographics, decision-makers, funding) while minimizing cost by querying premium providers only when cheaper ones fail.

Produces:Workflow
2

Primary Enrichment Pass

configure_enrich_step

Configure first enrichment step to query the primary provider (configurable — typically the cheapest provider with broad coverage, e.g. Apollo or Clearbit). Targets: company name, size, industry, tech stack, revenue band, key decision-makers. Outputs to account attributes. Mark fields successfully filled to inform downstream branching.

Produces:Step
3

Branch on Primary Coverage

configure_workflow_branch_step

Branch step: did the primary enrichment fill the required fields? If YES → skip to AI-fit-scoring. If NO (missing email format, missing decision-maker, or missing tech stack) → continue to secondary provider. Saves money by not paying for premium providers unless needed.

Produces:Step
4

Secondary Enrichment Pass

configure_enrich_step

Configure second enrichment step that fires only on the no-coverage branch. Query secondary provider (configurable — typically a specialist provider for the missing field type, e.g. ZoomInfo for decision-makers, BuiltWith for tech stack). Only fills fields the primary missed; doesn't re-query already-filled fields.

Produces:Step
5

AI Research Fallback

configure_ai_research_step

Configure AI research step that fires when secondary provider also fails to fill a key field. Tasks: scrape the company website, summarize what the company does, identify likely buyer personas from About/Team/Leadership pages, look up recent news for funding/hiring signals. Returns structured output (industry, ICP-fit description, 3-5 decision-maker names with titles). The Claygent-equivalent for the long tail where structured providers have no data.

Produces:Step
6

Update Account Record

configure_update_attribute_step

Configure the update step that writes all enriched data back to the Account record. Includes a metadata field 'enrichment_source_used' (primary / secondary / ai-research) so RevOps can audit cost per record. Also writes enrichment_confidence (high/medium/low based on which tier filled the data).

Produces:Step
7

Validate and Publish

publish_workflow

Validate the workflow DAG and publish for live execution. Set up a daily summary of enrichment-tier-usage so RevOps can monitor cost (e.g. '80% filled at primary tier, 15% at secondary, 5% needed AI research' = healthy cost profile).

Produces:Workflow

Ready to run this recipe?

Tell Blu what you need. It builds the segment, content, journey, and dashboard in minutes.

Sign up free

Hire the team that reclaims what your GTM stack is leaking.

You set the strategy. Agents run the plays. Seven AI agents across design, marketing, sales, and analytics. One customer context, tracked from first pixel to final dollar.

Start for free
Waterfall Account Enrichment Cascade | Intempt