Technology and infrastructure

Your technology stack determines whether your ABF operations run smoothly or become an operational nightmare. At small scale, you can get away with spreadsheets and manual processes. But as you grow, technology becomes the difference between a $50M facility that takes three people to manage and a $500M facility that takes the same three people with the right systems.

This topic covers what you actually need at each stage of growth, how to make build-vs-buy decisions, and how to avoid the common mistakes that cost originators money and credibility.

What systems you actually need

The core technology stack for ABF operations has four components:

System	What It Does	When You Need It
Loan Management System (LMS)	Origination, servicing, payment processing, collections	Day 1 of origination
Data Warehouse	Historical data storage, analytics, reporting	When lenders want cohort analysis
Reporting/Analytics	Borrowing base, covenant testing, performance reporting	First facility
Investor Portal	Self-service access for capital providers	$100M+ or multiple facilities

Your lender cares about one thing: can you produce accurate, timely data that matches what’s in your systems? If your monthly servicer report takes two weeks of manual work and still has errors, that’s operational risk they’ll price into your deal or decline to fund.

Note: Start with the end in mind. Even if you’re running on spreadsheets today, design your data capture as if you’ll need to produce institutional-quality reporting. Retrofitting data you didn’t capture is expensive.

Deep-dive topics

Each component of your technology stack warrants detailed attention:

Loan management systems - Core functionality, platform options by asset class, what lenders check during due diligence
Data and analytics infrastructure - Why you need both LMS and data warehouse, architecture, what to store, cloud platforms
Investor portals and integration - Portal options, report formats, integration requirements with payment processors, banks, credit bureaus, and counterparties
Build vs buy decisions - Decision framework, total cost comparison, vendor selection criteria, landscape overview
Security and compliance - SOC 2 requirements, security controls, privacy compliance, disaster recovery

Scaling considerations

Technology that works at $25M breaks at $250M. Know when to invest.

Volume triggers

Metric	Manual Process Breaks	Investment Required
Loan count	5,000-10,000	Automated tape generation
Monthly payments	10,000+	Payment processing automation
Daily draws	10+	Automated borrowing base
Facilities	3+	Multi-facility reporting
Investor reports	5+/month	Investor portal

Common scaling bottlenecks

Reporting speed: Monthly close that takes two weeks needs to be two days.

Solution: Automated data pipelines, pre-built report templates

Payment processing: Manual posting can’t handle volume.

Solution: Straight-through processing with exception handling

Reconciliation: Daily reconciliation becomes full-time job.

Solution: Automated reconciliation with exception-only review

Borrowing base: Weekly calculation takes all week.

Solution: Automated eligibility testing, real-time concentration monitoring

Staffing vs. automation

The rule of thumb: if a task takes more than one FTE of effort, consider automating. But automation has upfront cost and maintenance burden.

Task	When to Automate	Typical Investment
Tape generation	>10 hours/month	$20-50K
Covenant testing	>5 hours/month	$10-30K
Borrowing base	>8 hours/calculation	$30-75K
Investor reporting	>3 reports/month	$50-100K
Payment reconciliation	>2 hours/day	$30-75K

Multi-facility complexity

Once you have 3+ facilities with different capital providers, complexity multiplies:

Different eligibility criteria per facility
Different reporting formats and frequencies
Different covenant definitions
Potentially overlapping collateral pools

You need systems that can handle facility-specific logic without manual intervention.

Implementation roadmap

Phase 1: foundation (0-$100M AUM)

Systems:

LMS: Select and implement off-the-shelf solution
Data: Clean exports from LMS, well-organized spreadsheets
Reporting: Excel-based with documented templates
Investor access: Shared folder or basic portal

Cost: $50-150K implementation + $100-200K annual

Focus: Get accurate data, reliable reporting, documented processes. Don’t over-engineer.

Phase 2: automation ($100M-$500M AUM)

Systems:

Data warehouse: Cloud-based, automated data pipelines
Reporting: Automated tape generation, covenant testing
Analytics: BI tool with self-service dashboards
Investor portal: Third-party or basic custom

Cost: $100-300K implementation + $200-500K annual

Focus: Reduce manual effort, improve reliability, enable self-service. Start SOC 2 process.

Phase 3: scale ($500M+ AUM)

Systems:

Custom integrations: APIs to all counterparties
Advanced analytics: Predictive models, real-time monitoring
Full automation: Straight-through processing where possible
Enterprise portal: Full investor self-service

Cost: $300K-1M implementation + $500K-1M annual

Focus: Operational efficiency, institutional-quality infrastructure, full compliance posture.

Budget guidance

Plan to spend 1-3% of AUM annually on technology infrastructure:

AUM	Technology Budget Range
$50M	$500K-1.5M
$200M	$2-6M
$500M	$5-15M
$1B+	$10-30M

This includes systems, staff, compliance, and ongoing maintenance.

Common technology mistakes

Building too early

The mistake: Custom-building systems before you have product-market fit or sufficient volume.

Why it happens: Founders assume their needs are unique. They’re usually not.

The cost: $500K-2M spent on systems that don’t match your eventual business. 12-18 months lost.

The fix: Use off-the-shelf solutions until you hit genuine limitations. Custom builds should solve proven problems, not anticipated ones.

Underinvesting at scale

The mistake: Continuing with manual processes as volume grows.

Why it happens: The team that got you to $50M thinks they can get you to $200M the same way.

The cost: Operational errors that damage lender confidence. Staff burnout. Inability to take on new facilities.

The fix: Proactive investment when you see volume triggers approaching. Budget for technology in your growth plan.

Ignoring data quality

The mistake: Accepting “close enough” data and reconciliations.

Why it happens: Investigating discrepancies is time-consuming. Small variances seem immaterial.

The cost: Lenders lose confidence when your numbers don’t tie. Small errors compound into material misstatements.

The fix: Zero tolerance for unexplained variances. Investigate every discrepancy, fix root causes, document the issue and resolution.

Single points of failure

The mistake: Critical processes that depend on one person or one system with no backup.

Why it happens: Expertise concentrates. Building redundancy feels inefficient.

The cost: When that person leaves or that system fails, you’re unable to operate.

The fix: Document all critical processes. Cross-train staff. Ensure system backups are tested and restorable.

Poor documentation

The mistake: Systems and processes that only one person understands.

Why it happens: Documentation isn’t urgent until the person leaves.

The cost: New staff take months to become productive. Errors during transitions. Lender concerns about key-person risk.

The fix: Require documentation as part of system implementation. Runbooks for all critical processes. Regular documentation reviews.

Technology checklist by stage

First facility ($25-100M):

LMS selected and implemented with all required data fields
Payment processing connected and tested
Loan tape generation automated (or documented manual process)
Monthly reconciliation procedures documented
Backup procedures in place
Basic access controls implemented

Growth stage ($100-300M):

Data warehouse implemented with historical snapshots
Automated covenant testing
Investor portal or structured document sharing
Credit bureau reporting operational
SOC 2 Type II process started
Disaster recovery plan documented and tested

Scale ($300M+):

SOC 2 Type II certification obtained
Multi-facility reporting capability
API integrations with key counterparties
Real-time concentration monitoring
Automated borrowing base calculation
Full business continuity plan tested annually