RELAY

From Spreadsheets to Scale: How Azure and AI Automated Draws and Unlocked Scalable Growth          

RELAY an AI-powered software for property managers, managing construction and replacement reserve draws through spreadsheets, email threads, and scattered documents—leading to frequent lender rejections, funding delays, and significant cash flow strain. Allata designed and built a custom web platform powered by Azure and AI-assisted invoice processing to automate the full draw lifecycle. The platform eliminated manual package assembly and positioned the firm to scale its portfolio without proportionally increasing headcount.

OVERVIEW 

RELAY, an AI-powered software for property managers, supports an ownership group that acquires multifamily properties at scale, renovates them, and resells them. Managing lender-funded construction and replacement reserve draws across 40–50 properties—with thousands of invoices per month—had been a manual, error-prone process that constrained cash flow and limited growth. 

Allata built a custom web platform that automates draw assembly, validates documentation against lender requirements, and uses AI to process and categorize invoices at scale. Designed for multi-tenancy, the platform is now positioned as a commercial SaaS product under a newly formed entity, supporting operators across real estate and adjacent industries. 

THE CHALLENGE

The client’s asset managers compiled draw packages manually, pulling invoices, checks, lien waivers, and progress photos from spreadsheets, email threads, and disconnected file stores. Each package required a meticulous assembly to meet lender documentation standards. Lenders often returned packages due to missing or incorrect documentation were common, leading to draw approvals often taking 3–4 weeks, forcing the company to float construction costs out of pocket. When vendors went unpaid, they risked placing liens on properties or halting work entirely, compounding delays and adding legal complexity. Additionally, asset managers devoted a significant amount of time to repetitive document gathering and manual data entry across hundreds of invoices each month. 

Without a scalable system, adding new properties to the portfolio meant hiring more people to manage the same manual processes. The firm’s ability to invest in additional properties was directly constrained by both the administrative burden and the cash tied up in delayed reimbursements.

OUR SOLUTION 

Allata began with a six-week discovery phase to map the client’s draw processes end-to-end—understanding how operators prepare, submit, and track draws across both construction and replacement reserve workflows. This foundation informed the design and build of a custom web platform, delivered in approximately eight months of implementation followed by a market-readiness phase completed in January 2026. 

The platform intercepts vendor invoices sent via email using Azure Logic Apps and Communication Services. Azure Document Intelligence extracts invoice data through OCR, and Gen AI fills in missing fields or classifies line items against budget categories defined in loan agreements. Duplicate detection and business-rule validation catch errors before they reach a lender. When a draw is ready, the platform assembles the full package—invoices, checks, waivers, photos—and guides users through a checklist to confirm completeness before submission. 

Bulk invoice upload allows multi-invoice files to be split, processed, and routed automatically, with anomalies flagged for human review. Smarty address verification matches invoices to the correct properties. The platform tracks spending against draw thresholds and notifies asset managers when a property meets both timing and dollar criteria to submit—eliminating the guesswork of managing dozens of properties simultaneously.

  • AI-assisted invoice processing: OCR extraction augmented by Gen AI to infer missing fields and categorize line items against loan-specific budget categories 
  • Automated draw assembly: System compiles and validates full draw packages, enforcing documentation completeness before lender submission 
  • Proactive notifications: Alerts asset managers when draw threshold and timing criteria are met, enabling timely submissions 
  • Multi-tenant architecture: Platform designed from the ground up for commercialization as a SaaS product, supporting onboarding of external organizations across real estate and adjacent industries 

“We’re thrilled with what Allata delivered in just eight weeks. We’re a proud partner and are looking forward to continuing to expand our AI capabilities under the expert guidance of their team.”

Ryan Beaupre, President, RELAY

THE RESULT  

The custom platform went into production in early 2026, replacing the manual workflows that constrained the client’s operations. Asset managers now focus on anomalies and exceptions rather than assembling every draw package from scratch. Automated validation ensures documentation meets lender requirements before submission, directly reducing rates and the rework cycles. 

By compressing the time between invoice receipt and fund release, the platform reduces the period during which the firm must float construction costs from its own capital. That freed liquidity translates directly into capacity to acquire and renovate additional properties. The same team of approximately 12 people can now manage a much larger portfolio—targeting 100+ properties, 2x from the current—without proportional headcount increases.  

technology

Built on Microsoft Azure, combining deterministic business rules with Gen AI to balance automation speed with human oversight on ambiguous cases.  

Innovation starts with a conversation.

Fill out this email form and we’ll connect you with the right person for your needs.

Related Case Studies

A Leading National Environmental Services Provider

AI-Powered Waste Profiling Cuts Rework by Half and Accelerates Approvals for a Leading National Environmental Services Provider           

A leading national environmental services provider processed thousands of hazardous waste profiles monthly through a manual workflow that averaged 3.7 days per approval — and 5.5 days when rework was required. With 23% of profiles kicked back for corrections and accuracy hovering at 75–80%, delays frustrated customers, burdened agents, and increased regulatory risk. Allata embedded an AI engineering team alongside the client's domain experts to build an intelligent document processing and validation platform on AWS. The solution targets 98.5%+ extraction accuracy and a 50% reduction in downstream corrections from day one of production.
A leading Provider in Modular Space Solutions

Leader in Modular Space Builds AI Foundation in Eight Weeks  Enabling Enterprise-Wide Adoption           

A leading provider in modular space solutions faced mounting pressure as field and sales teams couldn't find critical product specs and procedures buried across 14+ disconnected systems and tangled SharePoint sites. After shutting down commercial ChatGPT access due to security concerns, the company needed a secure, enterprise-controlled AI solution that could deliver productivity gains without compromising corporate data. Allata deployed the Allata AI Accelerator Framework in an 8-week engagement, establishing a secure multi-model AI platform with two sales-focused personas that positioned the organization for broader AI adoption while eliminating security vulnerabilities.
Leading Biomedical Research Company

Accelerating Application Development: 4X Faster Front-End Delivery for Leading Biomedical Research Company          

A leading biomedical research company with legacy system limitations and overwhelmed IT resources partnered with Allata to accelerate critical application development. By deploying a team of four senior front-end developers with specialized Vue.js expertise, Allata delivered a fully functional geneticist application in just three months—a timeline that would have taken the client's internal team more than a year to achieve, representing a 4X improvement in development velocity.