Most of what ranks for “self-managing a rental” is either a property management company explaining why you cannot do it, or a listicle that stops at “screen your tenants and keep good records.” Neither tells you the actual job. Self-managing is not a mindset, it is an operating system: a fixed set of recurring tasks, a handful of hard dates, and a place to keep the paper. Once you can see the whole system, the only real decision is whether you want to run it or pay 8 to 10 percent of rent to someone who will.
This page lays out that system for a 1-to-10-unit portfolio, with the real hours and dollars at each size, and links to the detailed pages for each piece. I self-manage my own small portfolio from two time zones away and close the books on the 5th of each month, so the numbers here are the ones I actually run, not a brochure. The work splits into five jobs: money in, money out, the calendar, the people, and the paper. Get those five running and the unit count stops being scary.
The five jobs that are the whole system
Strip away the advice and self-management is five recurring jobs. Each one has a tool, a cadence, and a failure mode worth naming:
- Money in: rent collection and tracking. Pick a payment method, then record every payment against a current rent roll the day it lands. The method matters less than the tracking; how to collect rent covers the options, and a rent roll template is where it gets logged. The failure mode is a late payer you forget to chase until day 20.
- Money out: the books. Every expense logged to the right category, every month, so January is not a shoebox in April. A rental property accounting routine on a fixed day keeps it from piling up.
- The calendar: renewals and deadlines. Insurance, property tax installments, rental licenses, and lease end dates, each with a lead time. This is the job that breaks people, because the minutes are small and the penalty for a miss is not.
- The people: tenants and turnover. Routine communication, the renewal decision 60 to 90 days out, and the occasional move-out. Turnover is the spike that dominates everything else.
- The paper: documents that win disputes. Leases, inspection reports, deposit itemizations, and dated communication stored where you can find them. Documenting tenant interactions is the difference between winning and losing in small claims.
Job one: rent in, recorded, and chased
Collecting rent is the easy half. The half that protects you is the record: a rent roll where each unit shows what was charged, what was paid, and on what date, so you can see a delinquency the moment it starts rather than three weeks later. Set a grace period in the lease, define a late fee, and apply both the same way every time; when rent is actually late and how much to charge in late fees settle the two questions people get wrong. When a payment is short, you want an escalation ladder you can run without inventing it under stress, from a friendly text to a formal pay-or-quit notice.
The discipline is recording on the day, not the method. A payment you mean to log later is a payment you will argue about later.
Job two: a monthly close that takes ten minutes
The books fail in one specific way: they are easy until they are six months behind, and then they are a weekend you will not give them. The fix is a fixed-day routine. On the same date every month, reconcile the rent you received, log the month's expenses to Schedule E categories, and glance at cash flow. The ten-minute monthly close is the whole habit; I run mine on the 5th, once the grace period has settled and the month is final.
Categorize as you go and tax season stops being an event. The two numbers worth watching are net operating income and cash flow, which are not the same thing once a mortgage is involved; NOI vs cash flow walks through why a profitable building can still feel tight in the checking account. If you are starting from nothing, a rent roll plus a categorized expense sheet is a complete first system.
Job three: the calendar that actually breaks people
Here is the part the listicles skip. The doing of self-management scales linearly: ten units is about five times the bookkeeping of two. The remembering scales worse. Every unit adds dates that all come due on their own schedule: an insurance renewal, property tax installments, a license renewal, a lease end. Ten units across six buildings can carry more than 30 hard dates a year before anything goes wrong, and the failure modes are not symmetric. Drop a task and you lose an afternoon catching up. Drop a date and you eat a lapsed policy, a tenant rolling into a holdover at last year's rent, or a deposit returned past a statutory deadline, with real penalties in many states; read your state's statute for the exact window.
The fix is to move the dates out of your head and onto something with a lead time attached, so you see a renewal 60 days out instead of the week it lapses. The annual checklist, month by month turns the floating dates into a fixed rhythm. When self-managing collapses, it is almost never because the work got hard. It is because the remembering did.
Jobs four and five: people, turnover, and paper
Routine tenant work is cheap per item and expensive in aggregate, because every open thread is a loop you carry until it closes. The expensive event is turnover. Showings, screening, make-ready, inspections, and deposit accounting run 20 to 30 hours per turn and cost real money in vacancy and make-ready; what turnover really costs prices it out. Tenure is the lever that controls it, which is why the renewal decision 60 to 90 days before each lease end is the highest- value 20 minutes you spend all year.
The paper is what makes the rest defensible. A move-in and move-out inspection with dated photos, a deposit return with itemized deductions, and a written log of who said what and when are the documents that decide a dispute. Treat the security deposit paper trail as the model for everything: if it is not written down with a date, it did not happen.
What it costs at 2, 5, and 10 units
The system is the same at every size; the load is not. At two units a typical week is 30 to 40 minutes and the job fits inside the margins of a full-time schedule, where the real risk is forgetting a system you touch twice a month. At five units the weekly load is about an hour and a half, and the dates no longer fit in your head: across three buildings you carry 15 or so hard dates a year, which is where ad hoc tracking starts dropping things. At ten units a quiet week is about three hours, but with multi-year tenancies you run three or four turnovers a year, so several months hold one in progress; averaged out, ten units is roughly 20 hours a month, a part-time job that surges.
Set that against the alternative. On a $1,400 unit, a 10 percent management fee is $1,680 a year plus a lease-up fee of half to a full month's rent at each turnover. On ten such units that is roughly $16,800 a year before lease-up fees, against the 240 or so hours those units would cost you, which buys your time back at about $70 an hour. The full fee math and the honest answer on how many units you can realistically self-manage each get their own page. The short version: four to six units with a demanding job and an ad hoc system, ten to twelve with checklists, reminders, and a vendor bench.
Where the system lives
For a long time a spreadsheet holds all five jobs, and there is no shame in starting there. The spreadsheet stops being free at the point where it can no longer surface the calendar, because a tab cannot warn you that a policy lapses in 30 days. That is the seam where landlords move to software, and the honest question is which piece you are buying for.
That seam is why rents.ai exists. The fastest way to judge it is the read-only demo at app.rents.ai/try: a pre-loaded four-unit portfolio you can click through with no signup, with the rent roll, the categorized finances, the renewal calendar, and the Schedule E rollup already running so you can see the whole system in one place. What it will not do is the doing: it does not collect rent, screen tenants, message your tenants, or dispatch a vendor, because those stay your job either way. It keeps the record and the calendar; you keep the relationships. If you can run those five jobs on a spreadsheet you do not need it, and that is the right test to apply.