Calculates renewal probability using usage, NPS, tickets, and engagement metrics. Generates AI retention plan when risk is high.

How it works 23 steps

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

Data fetch

▼ company company-reader-hubspot

Fetch reader from hubspot

▼

Process (22 steps)

⚠ guard_missing_contract_date guard

Stop if company contract end date is missing

■ days_until_contract_end date_diff

Calculate days between now and company contract end date

■ contract_in_range all

Check if all conditions are true

⚠ guard_contract_window guard

Stop if contract in range equals

■ has_usage_data compare

Check if company product usage trend 90d exists

■ usage_score_computed math

Calculate: min(100, max(0, ({{step.company.product_usage_trend_90d}} + 20) / 40 * 100))

skip: step.has_usage_data equal false

■ usage_score coalesce

Use the first available value from: usage score computed, setting scoring fallback value

■ has_nps_data compare

Check if company latest nps score exists

■ nps_score_computed math

Calculate: min(100, max(0, {{step.company.latest_nps_score}} * 10))

skip: step.has_nps_data equal false

■ nps_score coalesce

Use the first available value from: nps score computed, setting scoring fallback value

■ has_ticket_data compare

Check if company support ticket count 90d exists

■ ticket_score_computed math

Calculate: 100 - (min({{step.company.support_ticket_count_90d}}, 10) / 10 * 100)

skip: step.has_ticket_data equal false

■ ticket_score coalesce

Use the first available value from: ticket score computed, setting scoring fallback value

■ has_engagement_data compare

Check if company days since last positive engagement exists

■ engagement_score_computed math

Calculate: 100 - (min({{step.company.days_since_last_positive_engagement}}, 90) / 90 * 100)

skip: step.has_engagement_data equal false

■ engagement_score coalesce

Use the first available value from: engagement score computed, setting scoring fallback value

■ renewal_probability weighted_score

Weighted score from: usage score (35%), nps score (25%), ticket score (20%), engagement score (20%)

■ renewal_summary to_summary

Format results as a summary card

■ risk_checklist to_checklist

Format results as an actionable checklist

■ should_run_ai all

Check if all conditions are true

■ ai_context pick

Keep only specific fields from

skip: step.should_run_ai equal false

✦ ai_analysis ai-prompt

AI generates a response from a custom prompt

skip: step.should_run_ai equal false

▼

Output

summary renewal_summary

checklist risk_checklist

ai_analysis ai_analysis

Settings 3 configurable

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

enable ai toggle = on

ai threshold number = 70

scoring fallback value number = 50