On June 25, 2026, MRI Software launched two new products on its MRI Agora platform. Both are available to MRI clients globally across commercial, multifamily, residential, social housing, local authority, and occupier sectors.

MRI Agora Intelligence

Agora Intelligence is a monitoring layer. It tracks workflows across leasing, finance, and operations every day. It then delivers role-specific briefings that classify each finding as a risk, trend, watch, or opportunity, ranked by business impact.

Three features stand out for finance teams:

• Explainable findings.

Each signal says what changed, why it matters, and what to do next.

• Traceability.

Signals link back to the underlying data and include confidence levels.

• Ask Agora.

A natural-language query feature lets users investigate a signal without leaving the platform.

According to Commercial Observer, Agora Intelligence uses a semantic layer that lets large language models interpret real estate data. That matters because real estate data is full of domain-specific structures, such as entities, leases, recoveries, and property hierarchies, that generic AI tools often misread.

MRI Agora Orchestrator

Agora Orchestrator is the action layer. It runs real-estate-specific agentic workflows using a catalog of pre-built agents that MRI has refined across its client base. Teams configure agents for their own operations. The Orchestrator handles routine execution automatically and routes complex cases to people.

Governance is built in. MRI highlights full audit trails, searchable and exportable logs, and role-level permission controls. As MRI CEO Patrick Ghilani put it: "Connecting AI agents to systems is becoming easier. The bigger challenge is ensuring those agents operate in ways organizations can trust."

That focus on trust is the right one for finance. Any automation that touches the books needs a clear record of what it did, when, and on whose authority.

A. The "Spam Folder & Attachment" Trap:

Manual email distribution reliance creates significant deliverability risks. Large PDF attachments, unverified sending domains, and bulk distribution lists routinely trigger enterprise spam filters. When an investor or property owner fails to receive a monthly financial statement, the resulting friction strains business relationships and creates redundant work.

B. Portal Fatigue and Low Engagement:

The industry's default response to email deliverability issues has been forcing stakeholders onto client portals. While portals are essential for static document archives, relying on them for time-sensitive notifications creates friction. Expecting stakeholders to proactively log into separate portals leads to low engagement and delayed approvals.

C. Static Data vs. Dynamic Operational Signals:

A standard 50-page monthly owner report provides historical context, but it rarely highlights immediate operational priorities. When a major commercial lease renewal hits a critical milestone or a capital expenditure threatens a budget, leadership needs an immediate, contextual signal rather than a retrospective month-end summary.

The last two rows are where finance teams should look closely. Bank reconciliation is the one month-end task that depends mainly

on data that doesn't originate in MRI.

• Bank data every day.

KriyaBalance takes bank activity from statement files such as BAI2 and MT940, plus direct bank connections (including Wells Fargo, TD, and RBC) and aggregators in North America, Europe, and Australia.

• Real estate matching logic.

Machine-learning matching handles complex, many-to-many transactions. When the engine isn't sure, it shows confidence scores and suggested matches, so people review only the exceptions.

• Bi-directional with PMX.

KriyaBalance reads from and writes to MRI Platform X, so it can create the required transactions directly in the MRI environment instead of leaving staff to key them.

• Controls built in.

Role-based access, multi-factor authentication, and full audit trails record every match and every manual decision.

• Built on Microsoft Azure

for security and scale across large, multi-entity portfolios.

The goal is simple: turn reconciliation from a 40-to-60-hour monthly task into a short daily review of exceptions.

Treasury 360 brings KriyaBalance together with KriyaVision (AI invoice processing) and KriyaCash (payments and cash management). For MRI Platform X teams, that adds:

• Three-way matchingbetween the bank, the ledger, and invoices.

• Cash flow forecastingwith fund-transfer recommendations across entities.

• Electronic payments with approval workflows, so cash leaves only after the right sign-offs.

Agora Intelligence is only as good as the data it reads. When bank activity is reconciled daily, the cash balances, receipts, and outstanding items inside PMX are current. Any cash-related signal that Agora raises then rests on matched, trustworthy numbers. In the other direction, exceptions that KriyaBalance can't resolve become clear, evidenced items for people to investigate, which is exactly what a governed month-end needs.

Use these steps to plan an AI-ready month-end without disrupting the close you run today.

1. Map the close.

List every month-end task, its owner, its system, and how long it takes. Mark which tasks start inside PMX and which depend on outside data, such as bank files, lockbox, card settlements, and invoices.

2. Ask MRI about the Agora catalog.

Confirm which pre-built agents and Intelligence signals are relevant to finance on your PMX deployment, and what licensing applies.

3. Count your bank accounts.

Record the number of accounts, banks, entities, and monthly transactions. This is the clearest measure of your reconciliation workload.

4. Measure the baseline.

Track hours spent on reconciliation, the number of unreconciled items at close, and how many days after month-end each account is signed off.

5. Fix the bank data path first.

Move from manual statement downloads to daily automated files or API feeds. Matching can't be faster than the data arriving.

6. Move from monthly to daily reconciliation.

Daily matching shrinks the close to a review of a few exceptions, rather than a month of backlog.

7. Define the human hand-off.

Set confidence thresholds, approval roles, and escalation paths for both Agora agents and reconciliation exceptions.

8. Keep one evidence trail.

Make sure auditors can see agent actions, automated matches, and manual decisions, with who did what and when.

9. Pilot on a slice.

Start with one region, one bank, or a group of entities. Compare hours and days-to-close against the baseline before scaling.

10. Review quarterly.

Agent catalogs and bank formats both change. Revisit thresholds, rules, and ownership every quarter.