Leasing chatbot for apartments vs hiring a leasing agent: the cost compared

A fully loaded leasing hire costs about $6,065 a month. A leasing chatbot for apartments costs $0.85 to $5.00 per unit per month on every rate any vendor publishes. That means the software is the cheaper monthly line for any portfolio under roughly 1,200 doors, and under about 4,070 doors if you buy at the entry tier. Those are not our estimates: the wage comes from the Bureau of Labor Statistics and the per-unit rates come from AppFolio's and Yardi's own pricing pages.

Last updated September 2026. Wage data from the BLS. Vendor rates verified 9 September 2026.

Almost every article comparing leasing AI to leasing staff picks a round salary number out of the air, compares it to a per-unit rate somebody else guessed, and calls it analysis. Both halves of that comparison are usually wrong, and they are wrong in the same direction: the salary is understated because it ignores what an employee actually costs an employer, and the software rate is overstated because it is copied from a listicle rather than a vendor.

So here is the arithmetic with both sides sourced. It is not complicated, and once you have it you can run your own portfolio through it in about a minute.

How much does a leasing agent cost?

Start with the wage. The BLS Occupational Employment and Wage Statistics program put the May 2025 median annual wage for real estate sales agents at $52,830. Narrow it to the real estate and rental and leasing industry, which is where apartment leasing consultants sit, and the median falls to $49,850. The spread is wide: the lowest 10 percent earned under $32,970 and the highest 10 percent over $123,590, because the top of that occupation is commission-driven sales rather than on-site leasing.

The wage is not the cost, though, and this is the part the round numbers miss. The BLS Employer Costs for Employee Compensation release for June 2026 found that for full-time private industry workers, total employer compensation cost averaged $54.00 per hour worked, of which wages and salaries were $36.97 and benefits were $17.03. Wages, in other words, are 68.5 percent of what an employee costs you. Benefits are the other 31.5 percent.

Divide the leasing wage by that ratio and you get the real number:

One caveat worth stating, because it is the sort of thing an honest comparison flags rather than buries: that 68.5 percent ratio is an all-occupations private industry average, not a leasing-specific one. Your own benefit load may run higher or lower. If you know your true burden rate, use it instead. If you do not, the BLS average is a far better starting point than pretending benefits cost nothing.

If you are pricing a property manager rather than a leasing consultant, the same method applies to a bigger number. The BLS median for property, real estate and community association managers was $69,990 in May 2025, which loads up to roughly $102,175 a year, or $8,515 a month.

What does a leasing chatbot for apartments cost?

Now the other side, and here the surprise is that most of the category refuses to answer. We checked the pricing URLs by hand on 9 September 2026. Respage returns HTTP 404. MRI Software returns 404. 365 Connect returns 404. BetterBot returns 403. LeaseHawk and Perq redirect their pricing paths to marketing pages. Nurture Boss serves a 404 page under an HTTP 200 status, which is worth knowing in general: a 200 on a pricing URL does not mean there is a price on it.

Every published rate in this category comes from a property management platform that treats the AI as a tier feature rather than a product. Those platforms have to print a rate card, so they do.

A method note, because it explains why you have probably never seen AppFolio's numbers: its visible pricing page shows Get a Quote on all three plan cards and zero dollar figures, while seven priced offers sit in the JSON-LD structured data at that same URL. The rate card is machine-readable and buyer-invisible at once. There is more detail on that, and on the rest of the field, in our breakdown of what an AI leasing agent actually costs.

The practical takeaway: every rate anyone publishes falls between $0.85 and $5.00 per unit per month. The $3 to $6 band that circulates for the specialist vendors cites no vendor source and runs off the top of the published range, which is two independent reasons to treat it as folklore.

Is a leasing chatbot cheaper than a leasing agent?

Yes for most portfolios, and by a wider margin than operators expect. One fully loaded leasing hire costs $6,065 a month. Divide that by each published per-unit rate and you get the door count at which the software bill finally equals one salary:

Read the bottom row again, because it is the conservative case. Even at the most expensive tier anyone publishes, you would need to be running more than 1,200 units before the AI line item costs as much as a single leasing employee. A 250-unit operator at AppFolio Core pays about $373 a month against $6,065 for the hire. The comparison is not close, and it is not close at any realistic portfolio size.

What can flip it is the minimum, not the rate. AppFolio requires 50 units plus a minimum spend it never quantifies. Yardi attaches $100 a month to Breeze and $400 to Breeze Premier at an identical $1.00 Residential rate, so at 100 units a Premier customer pays an effective $4.00 per unit, four times sticker. Divide the minimum by the per-unit rate and you get the door count where the floor stops binding: 100 units on Breeze, 400 on Premier. Ask for the minimum before you ask for the rate.

Can a chatbot replace a leasing agent?

No, and any vendor implying it can is overselling. The honest framing is coverage, not substitution. A leasing office open 40 hours a week covers 23.8 percent of a 168-hour week. Staffing the whole week would take 4.2 full-time people, roughly $305,651 a year in loaded cost. Nobody does that. The other 76.2 percent is simply dark.

That dark window is what the software is for. A renter shopping on a Saturday evening messages five communities and tours the ones that replied. Speed to lead is decided in the hours your office is closed, which is also when most renters actually search. Buying AI to replace a person you already employ is usually the wrong business case. Buying it to cover 128 hours a week you were never going to staff is the right one.

It is also worth separating the leasing conversation from the rest of what a leasing office does. Prospect messaging is one job. The internal side, the meeting scheduling, the drafted follow-up email, the filing and the document chasing, is a different job with its own kind of assistant and should not be lumped into the same decision.

How many units do you need for a leasing chatbot to be worth it?

Far fewer than the crossover table implies, because the crossover answers the wrong question. It tells you when software equals a salary. What decides the purchase is how many leases you are currently losing to silence.

Run it your way. Take your monthly inquiry count, estimate the share that arrives outside office hours, and multiply by your normal inquiry-to-lease rate and your average annualized rent. If a 200-unit community fields 120 inquiries a month, half arrive after hours, and one in forty of those would have leased, that is 1.5 leases a month you are not capturing. At $1,600 a month in rent, the annualized value of those leases dwarfs any per-unit rate on the table above. If the number comes out small, do not buy. That is the only calculation in this decision that is genuinely yours.

What does a leasing chatbot actually do?

The useful ones do three things and hand off the rest. They answer availability, rent, pet policy, parking, deposit and lease-term questions from your real data. They prescreen on the objective criteria you define, move-in date, budget, occupants and pets. Then they offer genuine open slots on your leasing calendar, confirm the tour in writing and send a reminder, because reminders are what move the no-show rate.

Everything past that should escalate to a person with the full thread attached: unusual lease terms, negotiation, anything an applicant disputes, and any question that edges toward a protected characteristic. On fair housing specifically, compliance is a configuration property rather than a software property. Ask every prospect the same objective questions, never act on protected characteristics, never let the AI approve or deny anyone, and have counsel review your question set before launch.

What this comparison does not tell you

Three things, and they matter enough to state plainly. First, the per-unit rates above buy an entire property management system with AI attached, not a standalone chatbot, so you are not comparing like with like unless you were going to buy the platform anyway. Second, the specialist vendors that publish nothing are generally deeper: voice, native two-way PMS sync, renewals. If you need those, the cheap platform tier will not do the job. Third, a per-unit meter grows your invoice every time you take on a property, whether or not the AI does more work, which is exactly the cost curve a growing operator should think twice about.

That last point is why flat per-account pricing exists. MessageAgent publishes three tiers at $79, $199 and $499 a month with no per-door meter and no minimum spend to negotiate, covering text, web chat, email and social DMs on one bill. It is planned rather than on sale today, and we are the conversation layer rather than a property management system, so if native Yardi sync or outbound calling is the requirement, one of the platform vendors fits you better. Full rate comparisons sit on our apartment chatbot pricing page, alongside the wider field on AI leasing agent software and the day-to-day view from a property management chatbot.

The short version

One leasing hire costs about $6,065 a month fully loaded. Published leasing chatbot rates run $0.85 to $5.00 per unit per month. Under roughly 1,200 doors the software is cheaper than the person at any published tier, and under 4,070 doors at the entry tier. Ask for the minimum before the rate, divide one by the other to find where the floor stops binding, and judge the purchase on the after-hours inquiries you are losing rather than on headcount you were never going to add.