de
Sales Decision CompassAI Sales Decision Operating System · V14
Connecting to the live product service
AI Sales Decision Operating System

The next step in turning decentralized sales data into traceability

Generate traceable, auditable sales action recommendations based on unified customer context, so teams know who to contact, when to follow up, which customers are more likely to close, and the opportunities and risks in the current funnel.

  1. 01
    Configure sales data sourceCreate on-site data warehouse connection records and enter CRM, email, meeting minutes and customer behavior data source identifiers.
  2. 02
    Generate customer contextUnify multi-source sales data by customer scope to form traceable customer context.
  3. 03
    Run agent analysisGenerate lead research, customer personas, email recommendations and funnel analysis in customer context.
  4. 04
    Review action recommendationsSales or management review recommendations, opportunities, and risks and manually approve, ignore, or mark high-impact recommendations for later processing.
  5. 05
    Export decision reportExport on-site reports for regular sales meetings or management reviews.
Customer context coverageNo customer context yetThe number of customers who have generated customer_context_record / the number of customers in customer_scope
Latest action recommendationsNo action suggestions yetCount the number of next_action_records in the last 7 days by generated_at
Suggestions awaiting approvalNo suggestions pending approvalstatus is the number of next_action_records pending approval
Number of opportunity tipsNo opportunity prompts yetThe number of opportunity_risk_records whose alert_type is opportunity
Number of risk warningsNo risk warning yetThe number of opportunity_risk_records whose alert_type is risk
Data quality tipsNo data quality tips yetThe number of records with non-empty quality_warnings in customer_context_record
BUSINESS INPUT

New project

Set up a sales decision workspace for a sales team or line of business, configuring data sources, account scopes, funnel stages and manual approval boundaries to generate traceable account context, agent analysis and next action recommendations.

Loading product-specific business fields.

Not saved
TRACEABLE OUTPUT

sales decision board

Results, sources, failures, approvals, and rollbacks are saved by project.

0 live runs
∅

No business results yet

The customer context has not been generated yet. Please configure the data connection first and provide four types of sales data source identifiers.

PRIVACY

Data use boundary

This product may process customer names, contacts, business opportunity records, email content, meeting minutes, customer behavior events and employee communication information. Please only access the authorized data range; do not upload sensitive personal information, account passwords, private keys or unauthorized customer information that are not related to sales decisions. All AI output must retain source references to facilitate auditing and deletion. Current 10 Role Review Supplement: Minimize processing of data by authorized customer scope, time windows, field whitelisting, and purpose restrictions. ; Sensitive personal information unrelated to sales decisions such as account passwords, API keys, tokens, private keys, payment information, government ID numbers, health information, home addresses, etc. are excluded by default. ; Email and meeting content should be summarized and quoted in fragments first, and source references should be retained to support auditing and deletion. ; Sensitive content in sensitive source identification, customer scope, approval person's email, source_references, email/meeting fragments, and exported reports is only visible to authorized roles, and is masked by default on the list page. ; Workspace deletion, deletion by customer, and deletion by source must clearly impact the scope of the customer context, agent analysis, recommendations, opportunities/risks, approval records, export reports, rollback snapshots, and audit logs. ;Project fields must not store clear text passwords, connection strings, API keys, tokens or private keys; if credentials are needed in the future, they must be saved by the server-side key management mechanism.

RETENTION

Data retention

In-site connection configuration metadata, customer context, agent analysis, action recommendations, approval logs, and export reports are retained by default for 180 days; administrators can delete by workspace or clean up earlier. Clear text keys and external system passwords should not be stored in project fields.

SAFETY

Human responsibility and rollback

The contact objects, follow-up time, deal possibilities, opportunities and risk tips generated by this product belong to AI sales decision support and are not factual guarantees, performance commitments or automatic execution instructions. Users must take action based on the customer's actual situation, contract status, compliance requirements and the judgment of the sales person in charge.

Rollback only means undoing the report version, suggestion status, approval status or workflow stage generated in this site, and restoring to the previous site record; since the product does not connect to external production systems to perform writeback, rollback does not mean undoing any changes in external CRM, email systems, data warehouses or customer communications.