Forecasts pipeline by month with weighted totals and AI narrative.

How it works 18 steps

Inspect the data fetches, transforms, gates, and output this recipe runs.

Data fetch

▼ deals deal-list-reader-hubspot

Fetch list reader from hubspot

▼

Process (17 steps)

⚠ check_deals guard

Stop if deals is empty

■ open_deals filter

Filter by condition

■ with_close_diff map

Extract close_date from each item

■ future_deals filter

Filter by condition

■ valid_deals filter

Filter by condition

■ with_weighted map

Extract from each item

■ with_stale_flag map

Extract last_activity_date from each item

■ with_expired_flag map

Extract close_date from each item

■ stale_deals filter

Filter by condition

■ expired_deals filter

Filter by condition

■ stale_count count

Count items in stale deals

■ expired_count count

Count items in expired deals

■ grouped_by_month group_by

Group by close_date

■ sorted_forecast sort

Sort by

■ forecast_table to_table

Format results as a data table

■ ai_context pick

Keep only specific fields from

skip: config.enable_ai equal false

✦ cfo_narrative ai-prompt

AI generates a response from a custom prompt (PII auto-protected)

skip: config.enable_ai equal false

▼

Output

table forecast_table

ai_analysis cfo_narrative

Settings 3 configurable

Configurable at install. Defaults shown — change them anytime in Recued.

enable ai toggle = on

stale threshold days number = 30

forecast horizon days number = 90