From lead intake to funded status. A sequential breakdown of how files move through signal extraction, fundability scoring, product matching, and lender routing — with clarity on where automation applies versus human review.
Each stage addresses a specific conversion requirement. Files progress through sequential validation, with automated processing at data-intensive stages and human review at decision points requiring judgment.
Initial file capture and normalization. Application data is structured into standardized fields regardless of submission format. Duplicate detection and basic completeness validation occur at this stage.
47+ fundability signals extracted from normalized data. Credit profile depth, cash flow patterns, time-in-business verification, industry classification, and collateral indicators are computed and weighted.
Extracted signals are processed through the fundability model. Output includes an approval probability score, confidence interval, and identification of limiting factors that constrain funding options.
Fundability score maps to eligible product categories. Sequencing logic determines optimal order of applications to maximize total funded amount while avoiding inquiry conflicts and timing collisions.
Matched products are routed to appropriate lenders based on current appetite, approval velocity, and historical success rates for similar profiles. Multi-lender submissions occur in parallel where appropriate.
Approved files proceed to documentation and funding. Status tracking continues through disbursement. Outcome data feeds back into model calibration for continuous improvement.
Funding timelines vary by product complexity, documentation requirements, and lender processes. Platform-assisted files demonstrate compressed timelines relative to traditional pathways due to pre-validated signals and optimized routing.