A system of record (Yardi, MRI) is the authoritative ledger of leases, rent, tenants, and history. A system of action is whatever chases the tenant, collects the document, sends the notice, and resolves the exception. Property management systems are built for the first job, not the second, and the gap between the two is where most operational risk lives.
Ask most commercial and retail property teams what runs their operations and the answer is almost always a platform like Yardi or MRI. That answer is correct, and it is also incomplete. Those platforms are a system of record: the single source of truth for leases, rent rolls, unit status, and financials. What they are not, by design, is a system of action: something that actually chases the tenant, collects the document, or resolves the exception. Confusing the two is where a lot of commercial real estate operations quietly go wrong.
That distinction is not a knock on Yardi, MRI, or any other property management system. They do what a system of record is supposed to do, and they do it well. The problem shows up when property teams expect the record to also be the doer, when the assumption is that because a lease expiration date lives in the PMS, someone or something is automatically acting on it. It usually is not. That gap between what is recorded and what is executed is the subject of this article.
What does a system of record do well?
A property management system earns its place at the center of the operation because it is authoritative. It holds the lease terms, the rent schedule, the unit and tenant roster, the accounting ledger, and the historical record of what happened and when. When a broker asks what the current rent roll looks like, or an auditor asks for the lease abstract on a given suite, the PMS is where that answer comes from, and it should be. Centralizing that data is what makes a portfolio auditable and financially sound in the first place.
Systems of record are also built for structure and permanence. They enforce a schema, they retain history, and they give every stakeholder (ownership, asset management, accounting, leasing) a shared source of truth to work from. None of that is in question. The issue is what happens next: a system of record can tell you that a certificate of insurance expired last Tuesday. It cannot, on its own, pick up the phone, send the follow-up, or make sure a replacement certificate is in hand before an incident turns that gap into a liability.
What does it leave undone?
Most property management systems were architected to store and report, not to initiate and follow through. That is not a flaw so much as a scope boundary, but it is a boundary that a lot of operators run past without noticing. The table makes the boundary concrete.
| The record says | The action that still has to happen | Who usually does it today |
|---|---|---|
| COI for Suite 210 expired Tuesday | Request to the broker with the exact requirement; two follow-ups; a call; a decision on the vendor due Thursday | The property manager, between showings |
| Renewal option lapses in 60 days | Draft the notice, confirm the notice address, send, log the acknowledgment, chase silence | An admin, from memory |
| Balance is 30 days past due | The collections conversation; the payment plan; knowing when a reminder should get firmer | Whoever has slack that week |
| Tenant request logged from the portal | Acknowledge in a human voice; dispatch an approved vendor; update on the ETA; tell the tenant it is done | The coordinator, if the coordinator is in |
| Sales report due on the 10th (retail) | Request, chase, validate the figure against history, enter it, flag the outlier | Nobody in particular, until month end |
The size of the ecosystem around these platforms is itself evidence of the gap: Yardi lists more than 450 active interface partners, and its own maintenance interface is documented as pulling lease and unit data and pushing work orders back into Voyager. The tools exist because the record is not the doer.
Where is the system-of-action gap?
The system-of-action gap is the space between "the data says this needs to happen" and "this actually happened." It is where tenant communication stalls, where compliance documents go uncollected past their deadline, and where lease events get noticed too late to act on cleanly. It exists in every portfolio that treats its PMS as the entire operation rather than the record layer underneath it, and it grows in direct proportion to portfolio size, because the volume of things that need doing scales faster than any one person's capacity to remember and execute them manually. The vendors are moving toward it from the platform side: Yardi's Virtuoso ships native agents for leasing, maintenance, and accounting workflows, and the strongest coverage is on tasks inside the platform, which is exactly where the unstructured email, the broker's phone call, and the lease clause in a PDF are not.
Red flag: if your team's answer to "what happens when the alert fires?" is a person's name rather than a process with a clock, the gap is open, however good the platform.
Closing that gap is exactly the kind of managed operation we built Premise to run. We sit on top of the system of record and act on what it surfaces: reaching out to tenants before an SLA slips, chasing and verifying compliance documents against the lease terms they are supposed to satisfy, and flagging lease events with enough lead time to act rather than react. The record stays authoritative (we do not replace it); we close the distance between what it says and what gets done, and we write back so the ledger stays the single record. You can see the specifics of what we handle across communication, compliance, and lease operations, and what integration with Yardi and MRI should actually mean.
How do record and action work together?
The right way to think about this is not record versus action; it is record and action, each doing the job it is built for. The system of record stays the ledger of truth: what the lease says, what the balance is, what the history shows. The system of action is what turns that ledger into follow-through: the outreach that happens, the document that gets collected and verified, the exception that gets escalated to a person instead of sitting unresolved. Property teams that only have the first are flying with excellent instruments and no autopilot; every course correction depends on someone noticing and acting manually.
This is also why AI-plus-human execution matters more than automation alone: volume needs an engine, but judgment calls (a tenant relationship that needs a real conversation, an exception that does not fit the standard playbook) still need a person who owns the outcome. We wrote about how that division of labor works in practice in AI plus human operators, and about the three ways to staff the action layer in managed operations vs. software vs. outsourcing. Both are worth reading if you are evaluating where the action layer should live.
Your PMS was never going to do the work; it was never built to. Owning the decisions instead of the paperwork starts with putting a real system of action underneath the system of record you already trust.
Frequently asked questions
What is the difference between a system of record and a system of action?
A system of record stores and reports: it holds the lease terms, the rent schedule, the tenant roster, and the history. A system of action initiates and follows through: it sends the request, chases the reply, verifies the document, escalates the exception, and tells the tenant it is done. Yardi and MRI are the first; the second is usually a person.
Is Yardi or MRI a system of action?
Mostly no, by design. Both fire alerts, hold documents, and now ship AI features for tasks inside the platform, but the outbound follow-through with tenants, brokers, and vendors still lands on the property team. Their integration programs are large (Yardi lists 450+ interface partners) precisely because other tools and services do the action layer.
What is the system-of-action gap?
The distance between "the data says this needs to happen" and "this happened." A certificate that expired last Tuesday, a renewal option that lapses in 60 days, a delinquent balance: the record shows each one. Somebody still has to chase, draft, call, and close, and that work is improvised in most portfolios.
Should I replace my property management system to close the gap?
No. Replacing the ledger moves the workflow problem to a new ledger and costs a year. Keep the system of record and put an action layer on top of it: a coordinator with a target and tooling, or a managed operation that works in your system under your policies with an SLA.
How should the action layer connect to the system of record?
At the round-trip level: work orders, compliance status, documents, lease events, contacts, and costs written back into the ledger, not held in a second system. If a property manager could ever ask "which system is right?", the integration is too shallow.