Manage more units without more property admin.
Every unit you add brings repair requests, tenant emails, contractor invoices, lease dates and a service-charge statement at year end. We look for the workflows where agents can triage, match and prepare that work against the lease and the ledger, with checks that stop a missed emergency or a wrong charge, and property managers keep every decision that affects a tenant's home or business.
Six workflows that grow with every unit under management.
These are the workflows we would examine first in a residential or commercial property manager, a housing company or an asset manager's operations team. They run through the property management system, shared inboxes, contractor portals, the accounting ledger and folders of scanned leases.
| Workflow | Manual today | AI-native version | Stays human | Verification check |
|---|---|---|---|---|
| Maintenance requests and contractor dispatch | Staff read each request, call the tenant back, work out urgency and responsibility, then find and brief a contractor. | An agent classifies the request, sets urgency by your rules, checks the lease for responsibility and prepares the work order for the right contractor. | Emergencies, safety issues, disputes about responsibility and work above a cost limit. | Safety and emergency keywords always go to a person at once. Work orders above the cost limit need approval. The contractor must be on the approved list for that trade. |
| Tenant enquiries | Questions about payments, keys, parking and house rules arrive by email and phone and wait in a shared inbox. | An agent answers routine questions from the lease, house rules and account data, and drafts replies for the rest. | Complaints, disputes between tenants and anything about rent changes or notices. | Answers must cite the lease clause or account record they come from. No commitments outside fixed templates. |
| Contractor invoices | Accounts staff match each invoice to a work order, check the rates and allocate the cost to the right property and cost type. | An agent matches invoice lines to the work order and the framework rates, and proposes the property and cost allocation. | Variances, disputed work and payment approval. | Invoice must match a closed work order within tolerance. Duplicate check on contractor, invoice number and amount. |
| Service-charge and operating-cost statements | Each year, staff collect costs, apply the allocation keys per building and prepare statements tenant by tenant. | An agent gathers costs from the ledger, applies the allocation keys from each lease and drafts the statements with a full working. | Final sign-off, disputed items and any judgment on what is chargeable. | Allocated costs must add up to total recoverable costs per building. Every charge traces back to a ledger entry and a lease clause. |
| Lease abstraction and key dates | Someone reads each lease to find rent reviews, indexation, break options and expiry, and keeps them in a spreadsheet. | An agent extracts key terms and dates from leases, with the page reference for each, and keeps the date calendar up to date. | Interpreting unclear clauses and any decision to serve or answer a notice. | Every extracted term links to its source page. A person confirms each abstract before it is relied on. |
| Rent arrears follow-up | Staff check the rent roll, send reminders and keep notes on payment arrangements by hand. | An agent matches payments, prepares reminders by your rules and summarizes each account's history. | Payment plans, hardship cases, legal steps and anything that could lead to losing a home or premises. | No reminder goes to an account with an agreed plan or an open dispute. Legal steps are never started by an agent. |
Six numbers that show whether it worked.
The Audit measures these before anything is built. The Pilot is measured against the same numbers.
| Metric | Definition |
|---|---|
| Admin hours per 100 units | People time spent on property administration per month, per 100 units managed. |
| Time to work order | Hours from a repair request to a contractor being instructed, by urgency level. |
| Emergency response time | Minutes from an emergency report to a person acting on it. |
| Invoice match rate | Share of contractor invoices matched to a work order with no manual correction. |
| Statement dispute rate | Service-charge statements queried or disputed, per 100 issued. |
| Missed key dates | Rent reviews, indexations, breaks or expiries not acted on in time. |
What can go wrong in property management, and the control for each.
Property work touches people's homes and businesses. Decisions about who gets a tenancy, rent changes, notices and legal steps stay with people. These are the failures we design against first.
| Failure mode | Control | How we'd detect it |
|---|---|---|
| Emergency treated as routine | Fixed keyword and category rules route gas, water, fire, electrical and security reports to a person at once, whatever the model thinks. | Review of every emergency against its response time, and a sample of routine requests checked for missed urgency. |
| Wrong party charged for a repair | Responsibility is checked against the lease. Unclear cases go to a property manager. | Recharges disputed by tenants, counted every month. |
| Service-charge statement wrong | Allocations must reconcile to the ledger per building. Every line has a working that a person signs off. | Reconciliation report per building before statements are issued. |
| Key lease date missed | Dates are extracted with source references, confirmed by a person and alerted well ahead. | Monthly report of upcoming and passed dates with their status. |
| Reminder or notice sent to the wrong tenant | Reminders only use templates. Accounts with plans or disputes are blocked. Formal notices are never sent by an agent. | Weekly check of reminders sent against plans and disputes. |
| Tenant personal data exposed | Agents see only the fields a task needs. Self-hosted, EU-hosted deployment is possible. | Access logs per agent and per task. |
Maintenance requests, from diagnostic to pilot.
Illustrative example — not a client result. The numbers below are assumptions chosen to show the method and the arithmetic. Your Audit replaces them with measured values.
| 01 | Diagnostic | Suppose a residential manager with 3,000 units picks "repair requests" on the Other / General Business track. The Snapshot might show a High leverage signal with Medium confidence, friction at the Demand/Input and Decision stages, and Level 3 (workflow redesign) as the indicated AI depth, because urgency and responsibility are decided case by case. |
| 02 | Audit | Suppose the Audit measures 900 requests a month at 12 minutes each from first contact to work order. That is 900 × 12 = 10,800 minutes, or 180 hours a month, about 1.1 FTE at 160 hours. It also finds that about 30% of requests are emergencies, above the cost limit or unclear on responsibility, and need a person. |
| 03 | Pilot | Suppose the 70% routine requests go through as prepared work orders, with a person checking a 10% sample for 4 minutes each. The 30% exceptions take 15 minutes each because they are the harder cases. The Pilot measures whether that holds. |
| Line | Assumption | Hours per month |
|---|---|---|
| Baseline | 900 requests × 12 min | 180.0 |
| Routine, after | 630 requests, 63 sampled × 4 min | 4.2 |
| Exceptions, after | 270 requests × 15 min | 67.5 |
| After, total | 4.2 + 67.5 | 71.7 |
| Difference | 180.0 − 71.7 (about 0.7 FTE) | 108.3 |
At an assumed loaded cost of €40 an hour, 108.3 hours is about €4,330 a month. The Audit would also measure time to work order and emergency response, because faster, correct dispatch matters to tenants more than admin hours do. It would also count the new costs: compute, property system integration and exception review. If the numbers don't hold, the recommendation is "don't proceed".
Where our proof comes from.
We'll be plain about this. Our production proof is in AdTech, not real estate. We have not yet run these property workflows for a client.
The architecture is the same one that runs in AdTech production: typed intake, deterministic checks before any action touches money or people, bounded retries, then a person. Classifying high volumes of incoming events and routing the risky ones to a person, as invalid-traffic detection does, is the same pattern as triaging repair requests. The case studies show that work, and five AI plugins built on this architecture are live at reviveadservermod.com.
That is why property engagements start with the free diagnostic and then the Audit. The Audit measures your own baseline first, so any decision rests on your numbers, not on ours.
Questions property managers ask first.
Will an agent decide which applicant gets a flat or unit?
No. Tenant selection affects access to housing and business premises, so the decision stays with people. An agent can check that an application is complete, but it does not score or rank applicants.
Can we trust an agent to read our leases?
Only with checks. Every extracted term links to the page it came from, and a person confirms each lease abstract before anyone relies on it. Unclear clauses are flagged, not guessed.
Do we need to change our property management software?
No. We work with the systems you have, through their APIs or exports. The Audit records which systems hold the data and where it is re-keyed.
Find out which property workflow is worth redesigning.
Start with the free diagnostic on the Other / General Business track. Move to the Audit only if the Snapshot points to something worth measuring.
| 01 | Free AI Operational Diagnostic | 10 steps, 8–12 minutes, Other / General Business track. Snapshot on screen and as a PDF. |
| 02 | Operational Leverage Audit, from $1,500 | Fixed fee, one workflow, 10–15 working days after data access. Measured baseline, benefit model with stated assumptions, go / no-go. |
| 03 | Pilot implementation, from $6,000 | Fixed scope. The workflow built and run in your environment, measured against the Audit baseline. Typically 4–8 weeks. |