Automated property performance tracking in Australia: what actually works
I've tried every way an Australian investor can track a property: the spreadsheet, the agent's portal, the accountant in July, the budgeting apps, the landlord software. Here's my honest read on where each one works and where it breaks, and how we built Kleev to close the gap that none of them do: statements that file themselves, gated by your approval.
The short version
- Australian investors track property performance five ways: spreadsheets, the agent's portal, the accountant at tax time, budgeting apps, and dedicated landlord software. Each covers a slice; none covers the whole.
- The common failure is the same everywhere: the data lives in monthly statements, and a human has to move it. That's the step that gets skipped, and skipped data is how a portfolio drifts for months without anyone noticing.
- We built Kleev to close exactly that gap: each property gets its own email address, statements sent there are read and categorised by AI, and every line waits for your one-tap approval before it touches your numbers.
- I'll also tell you where Kleev doesn't win: tenant management, rent collection and depreciation schedules belong to other tools, and I'd rather say so than pretend.
Before I built Kleev's property tracking I did what most Australian investors do: I kept a spreadsheet, skimmed the agent's monthly email, and found out how my property had actually performed once a year, from my accountant, months after any of it was actionable.
That lag bothered me more than the admin did. A property is usually the biggest leveraged position a person holds, and most owners can't answer "what did it cost me last month?" without an archaeology session. So here's my honest map of the options, including the ones we compete with, and where each genuinely wins.
Option one: the spreadsheet
The default, and I'll defend it further than most software founders will. A spreadsheet gives you full control and zero lock-in: every formula visible, every column shaped to your exact situation, at no cost. For one property and a simple loan it covers almost everything, and building it teaches you how your property works.
It breaks in three predictable places. Someone has to retype every lender statement, agent summary and bank line, month after month, and that someone is you.
Offset interest accrues daily on balance minus offset, which a monthly grid can only approximate. And formulas rot: a pasted-over cell or stale assumption sits wrong for months before anyone catches it.
We've written a whole honest comparison of spreadsheets and apps, and the punchline is that most property spreadsheets die of unentered months, not bad maths.
Option two: the property manager's portal
If an agent manages your property, they already run software, typically an agency platform of the PropertyMe or PropertyTree kind, and you get an owner portal or a monthly statement out of it. For what it covers, it's authoritative: rent collected, agent fees, bills the agent paid. That statement is the source document everything else should reconcile to.
But those platforms are built for the agency, not for you. The portal sees one agency's slice of one property: no loan, no offset, no expenses you paid directly, no council rates that skipped the agent, and no whole-portfolio view if your second property is managed by someone else.
Switch agencies and the history stays behind. The portal answers "what did the agent handle?", never "how is this property performing?".
Option three: the accountant, once a year
Your accountant produces the most accurate picture you'll ever get, exactly once, months after the year it describes. That's not a criticism; it's what tax compliance is for. But an annual reconciliation is useless for the decisions that come up in between: whether to fix the rate, raise the rent, sell, or fund the deposit on the next one.
There's also a cost angle worth saying plainly: an accountant handed a shoebox reconstructs your year at their hourly rate. An accountant handed a categorised year checks it. Whatever tracking you choose should aim to turn July into the second conversation.
Option four: generic budgeting and money apps
Budgeting apps solve the data-entry problem for everyday spending, usually via bank feeds, and if all you want is "how much did I spend on the rental?" they'll get you a category total.
What they lack is property as a concept. No yield, no equity or LVR, no interest-versus-principal split, no offset modelling, no statement grid that matches how a rental's year is actually structured, and no idea what an agent statement is.
The deeper problem is that a rental's most important documents never appear in your bank feed at all. The agent's statement arrives as a PDF, with the rent, fees and repairs netted into a single deposit your bank sees as one line. An app that only reads bank data reads a summary of a summary.
Option five: dedicated landlord software
There is real software built specifically for self-managing landlords. Tools like Landlord Studio do rental accounting properly: income and expense tracking per property, receipt capture, rent reminders, reports at tax time.
If you self-manage and want tenant workflows (rent collection, reminders, lease paperwork) in the same place as the bookkeeping, this category genuinely earns its keep, and I'd point you at it without hesitation for that use case.
Two caveats for Australian investors. Much of the category is built abroad: Stessa, one of the best-known names, supports US properties only as at August 2026, and even the tools that do serve Australia often need their tax reporting adapted to Australian rules rather than speaking them natively.
And these tools centre the tenancy, not the investment: the loan, the offset and your position across the whole portfolio usually sit outside the frame.
| Approach | What it genuinely does well | Where it falls short |
|---|---|---|
| Spreadsheet | Full control, zero cost, shaped exactly to you | −You are the data feed; offset maths and formulas rot |
| Agent portal / statements | Authoritative for rent, fees and agent-paid bills | −One agency's slice: no loan, no offset, no portfolio view |
| Accountant at tax time | The most accurate picture you'll get | −Arrives once a year, after every decision it could have informed |
| Generic money apps | Bank transactions arrive without typing | −No property concepts; agent statements never reach a bank feed |
| Dedicated landlord software | Real rental accounting plus tenant workflows | −Often built abroad; centres the tenancy, not loan or portfolio |
| Kleev | Statements file themselves, gated by your approval; loan, offset and portfolio in one model | −No tenant management or rent collection; AI features process data via Anthropic's API |
The five common approaches plus ours, honestly scored. Minus marks the shortfall of each row, including our own.
The gap all five share
Look at the failure column and it's the same failure wearing five outfits: the data lives in monthly statements, and a human has to move it.
The spreadsheet needs you to type it. The portal holds it but won't connect it. The accountant connects it once a year. The money app never receives it. The landlord software receives it if you enter it.
Whoever removes that human step, without removing the human's control over the result, wins this category. That's the sentence we built to.
How we automated it: every property gets an email address
Here's the part I'm genuinely proud of. Every property you add in Kleev gets its own dedicated email address. Give it to your agent as the statement delivery address, or set up an auto-forward from your own inbox, and the pipeline takes over from there.
When a statement lands, AI reads the attachment and first works out what it's looking at. A statement (money that moved) is parsed into individual rows, each dated, categorised and signed.
A bill (a council rates notice, a water invoice, an insurance renewal) is deliberately held apart as a supporting document instead, because if your agent pays that bill, the payment already appears inside the agent statement, and parsing both would double-count it.
If you pay a bill yourself, you opt its lines in explicitly. That one distinction is where most naive automation quietly corrupts a year of figures.
Then comes the part I'd call the feature, not the compromise: nothing touches your numbers until you approve it. Parsed rows queue in a review table where you can correct a date, an amount or a category, then approve or reject the lot in a tap.
Inbound email is untrusted input, and an automation that writes strangers' attachments straight into your financial records is not automation I'd accept for my own money. You stay the editor; the pipeline just stops being you.
Approved rows flow into the statement grid, the year laid out with categories down the side and months across the top, every cell traceable to its source.
The same model drives the loan and offset projections (we've shown what an offset actually saves elsewhere) and, come July, the EOFY tax pack collects the year's deductions into a schedule your accountant can check instead of reconstruct.
Where Kleev honestly doesn't win
- Tenant management. Kleev doesn't collect rent, chase arrears, or handle leases. If you self-manage and want those workflows, dedicated landlord software is the right tool and I'd tell you to use one.
- Depreciation schedules. A quantity surveyor's schedule regularly pays for itself, and no software, ours included, replaces one. Kleev tracks what the schedule gives you; it doesn't produce it.
- Bespoke modelling. A subdivision, a granny flat feasibility, your own exotic scenario: a spreadsheet's freedom still beats any app's structure, and plenty of sharp investors run both.
- Authoritative records. Your agent's statement and your lender's records remain the source documents. Kleev is the live working view that reconciles to them, not a replacement for keeping them.
And a privacy note I want to make precisely
Kleev's core promise is structural: your bank CSVs parse in the browser, the raw file is never uploaded, and there's no bank login and no screen-scraping, ever. But the AI features are a different mechanism and I won't blur the two.
When AI categorises transactions or reads a statement you emailed in, that data (descriptions, dates, amounts, statement rows) is sent to Anthropic's Claude API for processing.
Anthropic states it doesn't retain API inputs on its standard tier, but the data does transit their infrastructure. That's the honest trade behind the automation, and you should make it knowingly.
So what would I actually do?
If you have one property and enjoy the spreadsheet, keep it, sincerely, and spend your energy keeping it current. If you self-manage and the tenancy admin is the pain, buy landlord software.
If what you want is the performance picture (what the property earns, costs and is building, across the loan, the offset and the portfolio, kept current without you retyping statements) then that's the exact gap we built Kleev to close.
Add a property, copy its email address, and send it a statement. The first approval takes about a minute, and it's the last time a statement waits on your typing. Put your property's statements on autopilot in Kleev →
Important: this is general info, not advice
- Kleev describes your own data and does not give financial advice.
- Third-party tools are described from their published information as at August 2026 and at a general level only; their features and availability change, so check their own sites before relying on any of it.
- How you track a property doesn't change what's deductible or how it's taxed. For anything tax-shaped, use a registered tax agent and check current guidance at ato.gov.au.
Common questions
What are the options for tracking investment property performance in Australia?
Five, broadly: a spreadsheet you maintain yourself; your property manager's owner portal and monthly statements; your accountant reconciling everything once a year; a general budgeting app reading your bank transactions; or dedicated landlord software built around rental accounting. Each covers a different slice, and most investors end up running two or three at once because none of them alone shows rent, loan, offset and expenses in one live view.
Is a spreadsheet enough to track a rental property?
For one property with a simple loan, honestly, yes. A well-kept spreadsheet covers rent, expenses by category, the loan split and value. It breaks in predictable places: someone has to retype every statement, offset interest that accrues daily doesn't fit a monthly grid, and formulas quietly rot. Most property spreadsheets die of unentered months rather than bad maths.
Why isn't my property manager's portal enough?
The portal is authoritative for what it covers: rent collected, agent fees, and bills the agent paid. But it only sees one agency's slice. It has no view of your loan, your offset, expenses you paid directly, or a second property managed elsewhere, so it can't tell you the number that matters: what the property earns or costs you after everything.
How does Kleev automate property statement entry?
Every property you add in Kleev gets its own dedicated email address. Ask your agent to send statements there, or auto-forward them from your own inbox. AI reads the attachment, works out whether it's a statement or a bill (bills are held as supporting documents so an agent-paid bill never double-counts), extracts and categorises every line, and queues the result for your review. You can edit any row, then approve or reject in one tap. Nothing touches your numbers until you approve it.
Does Kleev send my property data anywhere?
Two different things happen and we'd rather be precise about both. Bank CSVs parse in your browser and the raw file is never uploaded; there's no bank login and no screen-scraping. AI features are different: they send data to Anthropic's Claude API for processing, including transaction descriptions, dates, amounts and statement rows. Anthropic states it doesn't retain API inputs on its standard tier, but the data does transit their infrastructure, and we won't pretend otherwise.