Customer story · September 2026

    From Hundreds of Hours to Ten

    How a national asset manager and receiver put agentic workflows into production, and took back half of every month.

    97%less time on monthly reporting
    1,000+properties with their own PMs, formats, and timelines
    4 weeksfrom discovery to a working proof on their own data

    The gap

    Everyone has AI.
    Very few have realized ROI.

    88%of real estate firms are piloting AI1
    5%have achieved all of their AI goals

    Why?

    Access was never the constraint. Every firm has a license, and someone on the team writing prompts that generally work. What almost nobody has is a system: AI wired into where the work actually happens, running every cycle on real data without a person babysitting every step.

    A chat window alone can’t track down the right documents, reconcile inconsistent formats, and produce an investor-ready report on deadline. That requires clean data pipelines, encoding your firm’s business logic, and handling the exceptions nobody wrote down.

    Orchestrating this system is hard, and it’s why most teams stall on their existing AI pilots.

    1 JLL, 2025 Global Real Estate Technology Survey, October 2025. Based on responses from 1,500+ senior decision-makers across 16 markets; 88% of investors, owners and landlords report piloting AI, while 5% report having achieved all program goals.

    The problem

    Every format imaginable.
    One monthly deadline.

    Who
    A national asset manager and receiver
    Portfolio
    Over a thousand properties across the country, each with its own property manager, formats, and timelines
    The work
    Collect property-level financials, review and audit the data, then produce reports for investors and the courts, every month and quarter

    That meant chasing property managers for documents, then running separate workflows per audience to produce the same reports on repeat, capping how fast the business could grow.

    300hours a month on reporting
    16,000+reports produced a year
    200+pages rebuilt by hand each cycle

    The solution

    Two agentic systems working together

    We embedded with the client’s team to map their processes, bottlenecks, and business logic, then built parallel agentic systems.

    Document collection

    Watches Outlook and SharePoint, identifies the required documents from each property manager, categorizes them, flags discrepancies, and drafts standard notifications as properties mature through their holds.

    Reporting

    Ingests monthly financials, extracts the data, and populates investor and court-ordered templates. A human-in-the-loop step lets analysts verify against source numbers before anything ships externally.

    Enabling the team

    Beyond the workflows, we helped the firm invest in its own team. Monthly workshops helped everyone get more out of AI, beyond the workflows we automated.

    • FoundationsPrompting & model selection
    • AdvancedDeterminism vs inference
    • Build your ownSkills & agents

    The same reports, before and after

    0102030405
    By handManual work

    Analysts gather every required property doc

    Copy and paste figures and pages by hand

    Rebuild the report from scratch each cycle

    Email out for review, re-explain every change

    Repeat for the next property

    With agentsAutomated work

    AI watches SharePoint and inboxes for new files

    Extracts figures and pages, cites all sources

    Assembles the report, logs each change

    Briefs the reviewer with a change summary

    A human reviews and approves

    Sign-off stays with a person. The agents never ship externally without approval.

    The value

    2 weeks back, every month

    97%reduction in hours spent on reporting
    300hours before
    10hours after

    The reporting that used to consume the back half of every month now runs in about a day. The team recaptures that capacity to spend on the deals and decisions that grow the firm.

    Two key takeaways

    Business context is key

    Standing up a model is the easy part. The hard part is encoding how a firm actually reasons through its work e.g. the exceptions, the thresholds, and judgment calls made from experience. That logic is also where the value sits, because it’s what makes the outputs usable in the real world.

    Start where the work repeats

    Chasing documents and aggregating data is the easiest work to eliminate with AI. The agent gets you to 80–90% done, a person takes it across the finish line, and the recaptured hours go to higher value work.

    Next: the client plans to expand with Ollo into budget variance reporting and portfolio alerts for critical dates.

    About Ollo

    We’re not software you manage. We’re a forward-deployed team that builds bespoke workflows around your data and processes: your buy box, underwriting logic, formats, and thresholds, encoded into an engine that’s yours.

    founding team
    40+ yrscombined AI / ML experience
    Decadesof CRE experience

    We prove the work before you commit.

    We build one of your highest-cost workflows on your real data at no cost. All it takes from your side:

    • Discovery workshop60 minUnderstand your process and system landscape, and define a scope and success criteria for the proof.
    • Kickoff30 minConfirm success criteria with the executive sponsor, identify blockers, get necessary IT access, and present a formal project plan.
    • Weekly check-ins30 minReview progress on what we’ve built and iterate collaboratively as it takes shape.
    • Working sessionsAs neededOne or two deeper sessions with the person who owns the process, so we capture the edge cases and judgment calls that aren’t written down.

    Your data never trains a model and never leaves the infrastructure we agree on. After the proof, we productionize, run, and maintain each workflow.

    Get started

    See it running on your own data.

    First workflow live in under 4 weeks. No commitment until you’ve seen it work.

    Talk to us