For most agents and small brokerages, buying leads is the cheapest way to start and the most expensive way to scale, a real estate marketing agency is usually the cheapest path per closing between roughly $8,000 and $40,000 in monthly marketing budget, and an in-house team only wins above about $25,000–$30,000 a month in media spend or roughly 15+ producing agents. In hard numbers based on industry benchmarks: portal referral programs (Zillow Flex, Realtor.com ReadyConnect/Opcity-style) cost 30–40% of your gross commission per closing with zero upfront risk; pay-per-lead vendors run $2,000–$4,500 in media per closing at typical 0.8–2% close rates; a competent agency plus $5,000–$10,000/month in ad spend lands at $1,500–$4,000 per closing once campaigns mature at month 3–6; a full in-house team costs $12,000–$20,000/month fully loaded before a dollar of media and takes 4–9 months to ramp. The reason the ranking flips at different volumes is that each model has a different fixed-cost floor and a different variable rate. Referral fees are 100% variable — you pay only on closings, so they never bankrupt you, but they also cap your margin forever at 60–70% of GCI and you never own the pipeline. Agencies are semi-fixed: a retainer of $1,500–$8,000/month in the US market (or 10–20% of managed ad spend) plus media, with the strategy, creative and tracking amortized across your whole database. In-house is almost entirely fixed: a marketing manager at $60,000–$95,000, a paid-media buyer at $70,000–$110,000 and a video editor at $45,000–$70,000, each loaded 25–30% with taxes and benefits, whether you close 2 deals or 20. This guide gives you the arithmetic, not the sales pitch: what each model really costs, the close rates you should model, the break-even volumes where the answer changes, and the two variables — speed to lead and database ownership — that move cost per closing more than the choice of vendor ever will.
What does each option actually cost per closing?
Start with the only metric that matters: total marketing cost divided by closed transactions in the same 90–180 day window. Using a $400,000 median sale price and a 2.5% listing or buyer side, one closing produces about $10,000 in gross commission income (GCI).
Buying leads via referral (Zillow Flex, Realtor.com ReadyConnect, broker-to-broker networks): you pay 30–40% of GCI at closing, so $3,000–$4,000 per deal. Upfront cost: zero. Effective cost per closing: $3,000–$4,000, plus the agent hours burned on the 90%+ that never transact.
Buying leads pay-per-lead (Realtor.com Connections Plus by ZIP, Ylopo/BoldLeads-style programs, Facebook lead vendors): $15–$60 per buyer lead and $40–$150 per seller lead in most US markets. At a realistic 0.8–2% close rate on non-exclusive portal traffic, you need 50–125 leads per closing, or $2,000–$4,500 in media — and you often sign 6–12 month contracts at $1,000–$3,000/month minimums.
Real estate marketing agency: $1,500–$8,000/month retainer (or 10–20% of ad spend) plus $5,000–$10,000/month media. Owned campaigns with proper qualification typically convert 2.5–5% lead-to-closing because leads are exclusive and nurtured. At $8,000–$15,000 total monthly cost and 4–7 closings, that's $1,500–$4,000 per closing after ramp.
In-house: $12,000–$20,000/month fully loaded for a 2–3 person team, plus media and a $3,000–$15,000/year tech stack. Below 5 closings/month attributable to marketing, cost per closing exceeds $4,000 — worse than simply paying referral fees.
What does a real estate marketing agency actually charge, and what should be included?
In the US, real estate marketing agency pricing clusters into three tiers. Boutique/solo-operator: $1,000–$2,500/month, usually ad management only. Full-service performance agency: $3,000–$8,000/month retainer, or a hybrid of $1,500 base plus 10–15% of managed ad spend. Enterprise/developer-side: $10,000–$30,000/month, typically including brand, video production and a launch calendar for a project.
The retainer is only worth it if it buys you things you cannot buy per-hour. A defensible scope includes: offer and positioning strategy (which zones, which price band, buyer vs seller); paid media across Meta, Google Search and Performance Max, plus YouTube/TikTok where relevant; landing pages and forms that feed your CRM directly rather than an inbox; 8–15 pieces of short-form video per month, because listing and neighborhood video is what keeps CPMs down; full attribution — which campaign produced which closing, not which campaign produced which click; and lead response infrastructure.
Red flags that predict overpaying: no named media buyer, reporting that stops at cost per lead, ad accounts owned by the agency instead of you, no access to your own pixel or conversion data, and 12-month contracts with no 60–90 day exit. Insist on owning the ad accounts, the Business Manager, the domain, the CRM data and the creative files. That single clause is the difference between renting demand and building an asset you keep when the relationship ends. At Growth Estate we structure the Estate Funnel around exactly that: strategy, paid media, video and an AI that answers and qualifies every lead in under five seconds — all running inside assets the client owns.
When does building an in-house real estate marketing team beat hiring an agency?
In-house wins on three conditions, and you need at least two of them. First, media volume: once you consistently spend more than $25,000–$30,000/month, a 15% agency fee equals $45,000–$54,000/year — enough to hire a dedicated media buyer who does nothing but your accounts. Second, headcount: a brokerage with 15+ producing agents generates enough listing inventory and internal content demand that a full-time marketer stays busy; below that, they idle at 40–60% utilization. Third, proprietary process: developers with multi-phase launches and brokerages with a distinctive coaching system get compounding returns from institutional knowledge that no external partner accumulates as fast.
The real cost of in-house is rarely the salary. Budget the fully loaded number: a marketing manager at $60,000–$95,000 becomes $78,000–$124,000 with payroll taxes, benefits and equipment at a 25–30% load. Add software the agency was absorbing — a CRM at $60–$120 per user/month (Follow Up Boss, Sierra Interactive, kvCORE-class platforms start at $500+/month at team level), design and video tools, call tracking, scheduling, reporting. Then add the two invisible line items: 60–120 days of hiring and 3–6 months of ramp during which output is 40–70% of steady state, and key-person risk — when your one marketer resigns, lead flow stops in about three weeks.
The honest rule: in-house buys control and speed of iteration; agencies buy pattern recognition across dozens of markets and campaigns. If you have never run profitable paid media yourself, hiring your first in-house marketer means paying someone to learn on your money.
Why do bought leads look cheap per lead and expensive per closing?
Because you are not buying a lead, you are buying a share of an intent event that three to six other agents also bought. Non-exclusive portal leads are, by design, distributed. The consumer submitted one form and gets five calls, so your conversion is a race, not a relationship.
The math is brutal but predictable. Widely cited lead-response research shows that contacting an inbound web lead within five minutes dramatically increases the odds of qualifying them versus contacting them 30 minutes later — the multiple commonly quoted is over 20x — and yet a large share of internet leads are never contacted at all. Industry benchmarks put average agent response time on purchased leads at hours, not minutes, and typical close rates on non-exclusive portal leads at 0.5–2%.
Run it: 100 leads at $35 each is $3,500. At 1.2% you get 1.2 closings, roughly $2,900 in media per closing, plus roughly 25–40 hours of agent calling. Add a 30–35% referral fee model on top for Flex-type inventory and you are at $3,000–$4,000 of pure margin surrender per deal. On a $10,000 GCI split 70/30 with the brokerage, the agent nets $7,000, pays $3,500 in referral, and keeps $3,500 — before taxes, car, and the cost of showing 14 houses.
Bought leads are still the right answer in two cases: you are new, have no database and cannot risk fixed costs; or you have excess agent capacity and want to convert idle hours into deals. They are the wrong answer as a long-term strategy because the cost never declines with scale and the asset never becomes yours.
How do you calculate your own cost per closing in 10 minutes?
Skip the vanity dashboards and build a five-line model in a spreadsheet. You need: monthly marketing cost (media + retainer or salaries + tech), leads generated, contact rate, appointment rate, and closings, measured on a lagged window because real estate has a 45–120 day sales cycle for buyers and 60–180 days for sellers.
The formula chain: Cost per lead = spend ÷ leads. Cost per appointment = spend ÷ appointments held. Cost per closing = spend ÷ closings, with spend taken from the period 60–120 days before the closings. Then compare against GCI: your marketing cost should land at 10–20% of GCI for a healthy owned-channel operation, versus the 30–40% you surrender on referral inventory.
Benchmarks to calibrate against, per industry ranges: Meta lead ads for buyer interest in US metros run $8–$45 per lead; seller/home-valuation leads run $25–$120; Google Search on high-intent terms like 'homes for sale in [neighborhood]' costs $2–$8 per click and $40–$200 per lead, but converts 2–4x better than social. Contact rate on your own leads should exceed 60–70% with automated instant response; appointment set rate 15–30% of contacted; appointment-to-close 20–35% for sellers, 10–20% for buyers.
If you only track cost per lead, every vendor looks great. Track cost per held appointment and cost per closing and the ranking usually inverts: the $12 lead source that never answers the phone becomes more expensive than the $90 lead source that books.
Why does speed to lead decide the winner more than the channel does?
You can run identical campaigns with identical budgets and get a 3x difference in cost per closing purely from response time and follow-up depth. This is the single largest controllable variable in the entire model, and it is where most agency-versus-in-house debates are misdirected.
Three numbers explain it. First, response speed: leads contacted in the first five minutes qualify at multiples of those contacted after 30 minutes, and the curve collapses after the first hour. Second, contact attempts: most agents stop after 1–2 attempts, while benchmark data across sales sectors puts the majority of contacts at attempt 4 through 8, across mixed channels. Third, channel: WhatsApp and SMS response rates run far above cold-call pickup and email open rates in most Spanish-speaking and increasingly US markets — a lead who ignores three calls will answer a message in two minutes.
This is why an AI responder changes the arithmetic more than switching ad platforms does. If your instant-response layer replies in under five seconds on WhatsApp, asks the four qualifying questions (budget, timeline, financing, zone), and books the appointment into a calendar, your contact rate can move from 30–40% to 60–80% and your cost per held appointment drops proportionally — without spending another dollar on media. Agency, in-house or purchased leads, this layer sits underneath all three and multiplies whatever you already have.
Practical minimum stack: a CRM with lead routing, an instant auto-response on the lead's preferred channel, a 7–14 day cadence of 8–12 touches, and long-term nurture for the 60–70% of leads whose timeline is 6–18 months out.
What compliance rules change the math for housing ads in the US and EU?
Housing advertising is a legally restricted category, and the restrictions have direct budget consequences — so model them before you compare vendors. In the United States, the Fair Housing Act prohibits discriminatory advertising based on protected classes, and following the 2019 settlement between Meta and civil rights organizations plus the 2022 US Department of Justice settlement, housing ads on Meta must run in the Special Ad Category. Practically that means: no targeting by age, gender or ZIP code; a minimum geographic radius (commonly 15 miles); severely limited detailed-interest targeting; and algorithmic variance reduction applied to delivery. The consequence is that housing campaigns typically carry higher CPMs and require creative and offer to do the targeting work that audience settings used to do.
In the EU, GDPR governs the lead capture itself: you need a lawful basis for processing, clear consent for marketing communications, a privacy notice at the point of collection, and honored deletion and access rights. The EU AI Act adds transparency obligations for AI systems that interact with people — users should be able to tell they are talking to an AI, not a human agent. On US phone and text outreach, the TCPA regime governs automated calls and texts to mobile numbers, including consent requirements and revocation handling.
The budget impact: expect housing CPMs 15–40% above unrestricted verticals, plan for lower match rates on lookalikes and retargeting, and require any agency or vendor to demonstrate compliant setup in writing. This section is general information, not legal advice — confirm your specific setup with your own counsel or compliance team before launching, especially on automated outreach and AI disclosure.
What is the cheapest realistic setup for your production level?
Match the model to your volume rather than your ambition.
Under 12 transactions a year (solo agent, thin database): do not hire an agency and do not hire in-house. Buy referral-fee inventory where you pay only on closings, work your sphere, and invest your money in a CRM ($60–$120/month) and an instant-response setup. Your marketing budget is time. Target cost per closing: 30–35% of GCI, but with zero downside risk.
12–50 transactions a year (established agent or small team, 2–8 people): this is the agency sweet spot. A $3,000–$5,000/month retainer plus $5,000–$8,000/month media gives you exclusive leads, real creative and attribution you can audit, typically producing 3–7 additional closings a month after a 60–90 day ramp — roughly $1,500–$4,000 per closing, trending down as retargeting pools and your database compound. Keep 20–30% of budget on brand and database reactivation, not just cold traffic.
50+ transactions a year, or $25,000+/month media, or 15+ agents: build a hybrid. Hire one in-house owner of the system — a marketing manager who controls the CRM, the data and the brand — and keep an agency or specialist contractors for paid media and video production, where the learning curve is steepest and turnover hurts most. Full in-house makes sense past roughly $40,000–$50,000/month in spend.
Developers with multi-phase launches: budget 1.5–3% of sales value on marketing per phase, front-loaded, and insist on lead-level attribution to unit sold before phase two pricing is set.
Frequently asked questions
Per closing, an agency is usually cheaper once campaigns mature: $1,500–$4,000 per closing versus $3,000–$4,000 for referral-fee programs charging 30–40% of GCI and $2,000–$4,500 in media for pay-per-lead sources converting at 0.8–2%. But the agency carries fixed monthly cost ($1,500–$8,000 retainer plus media) and a 60–90 day ramp, while bought leads cost nothing until they produce. Below roughly one closing a month from marketing, buying leads is the lower-risk answer; above three to four closings a month, the agency almost always costs less per deal and builds an asset you keep.