Summary: AI voice agents don’t replace a real estate ISA — they replace the specific parts of the job that cause its 60–80% annual turnover rate. The winning model in 2026 is hybrid: AI handles after-hours leads, spike volume, and repetitive follow-up, while a human ISA focuses on nuanced qualification and relationship-building, producing more booked appointments than either working alone.
The ISA seat is one of the most discussed and least understood roles in residential real estate. Team leads know they need one. Some teams have burned through three of them. Others have never made the math work. And in the last 18 months, a new question has landed on top of the old ones: does an AI voice agent do the same job for a fraction of the cost?
The short answer is that AI does some of the ISA job, dramatically better and cheaper than a human can. It does not do all of it. The teams making the biggest gains in 2026 aren’t picking between AI and a human ISA. They’re running both, with each covering the specific work the other can’t.
This article is a guide to the whole picture. What the ISA seat actually does. What it actually costs. When to hire one. Why the industry’s 80% turnover rate exists. Where AI genuinely fits in. And how the hybrid model works when you build it right.
If you’re a team lead trying to decide whether to hire your first ISA, whether to keep the one you have, or whether to add AI to your current team, this is meant to give you the framework to make that call.
What a real estate ISA actually does
A real estate ISA (Inside Sales Agent) is a phone-based salesperson who works the top of your team’s funnel. They call new leads back in minutes, qualify them, follow up on your database, prospect old leads, and book appointments for your producing agents.
The role migrated into real estate from software sales in the mid-2010s, where the split between “inside” (phone-based) and “outside” (in-person) sales was well-established. In a real estate team, the split works the same way. The ISA handles every conversation from first inquiry through booked appointment. The producing agent (sometimes called an OSA, or outside sales agent) takes over from the appointment forward. The ISA never shows a house. The agent never dials cold leads for the first time.
The reason this split exists is economic. A producing agent’s time is worth a lot per hour once they’re in front of a live buyer. Every hour they spend dialing new leads is an hour they’re not negotiating, showing, or closing. Handing the phone work to an ISA frees the producer to do what they’re highest-paid for. Done right, this raises the whole team’s GCI without adding a licensed producer to the roster.
A typical ISA workflow looks like this:
New leads come in from Zillow, Realtor.com, Facebook Lead Forms, the team’s IDX website, and paid ad landing pages. The ISA is the first human contact. Their job is to reach the lead as fast as possible (industry benchmarks put the ideal response inside five minutes), qualify them using a framework like LPMAMA (Location, Price, Motivation, Agent, Mortgage, Appointment), and either book a showing or a buyer consultation on a producing agent’s calendar.
Between new leads, the ISA works the follow-up queue. This is the database of leads who didn’t book on first contact and are now sitting in Follow Up Boss, Sierra Interactive, kvCORE, or BoomTown at various stages of nurture. Some are three days old and warm. Some are 90 days old and need a check-in. A productive ISA makes 75 to 100 dials per day covering this mix, plus outbound cold prospecting for teams that run it (which uses power dialers like Mojo, RedX, Vulcan7, or PhoneBurner to hit 200+ dials in a shift).
The ISA does not close deals. They open them and hand them off. That handoff is the entire product they produce.
The economics of the seat
The salary line is the smallest surprise in an ISA hire. The real cost sits under it in ramp time, management overhead, tooling, and turnover.
Base salary and OTE. The industry standard OTE (on-target earnings) for a real estate ISA sits between $55,000 and $65,000 per year, split between a modest base and performance-based bonuses. Base salary usually lands between $30,000 and $45,000. The bonus structure varies by team but generally follows one of two models: 5% to 10% of GCI (gross commission income) on deals that close from ISA-booked appointments, or a flat per-appointment bonus of $50 to $150 for appointments held (not just set, because paying on “set” produces low-quality appointments the team lead later regrets).
Fully loaded first-year cost. Once you add employer taxes, benefits, tools (CRM seat, dialer subscription, phone number, call recording software), training time, and management overhead, the fully loaded first-year cost for an in-house ISA lands closer to $75,000 to $85,000. Most of the delta between OTE and loaded cost is the ramp period, when the ISA is being paid but not yet producing full-throughput results.
Virtual ISA services. Companies like MyOutDesk, Aiva, and various offshore providers offer virtual ISA services where you rent a trained caller plus their management layer for a monthly fee. Rates run from $720 per month for part-time coverage (roughly 2 hours per day) to $1,988 per month for full-time coverage. Annual cost lands between $8,640 and $23,860. That’s substantially less than an in-house hire, but you trade away control of scripts, voice, and quality of feedback loops. Virtual ISAs also don’t build institutional knowledge of your inventory or market the way an in-house hire does over time.
AI voice agents. The newest category. AI voice agents like Callingly answer inbound leads in seconds, qualify them using scripts the team owns, and warm-transfer to a live rep or book an appointment directly. Pricing runs from $100 to $750 per month for typical team volumes, with per-call overages that scale with usage. AI voice agents don’t do everything an ISA does (more on that below), but for the specific tasks they cover, they cover them 24/7 without ramp time or turnover.
Cost per held appointment. The right way to measure any of these options isn’t the sticker price. It’s cost per held appointment (an appointment the buyer actually attends, not just one on the calendar). A $75,000 in-house ISA who sets 80 appointments a year with a 60% show rate produces 48 held appointments at a fully-loaded cost of roughly $1,560 per held appointment. A virtual ISA at $18,000 producing 120 held appointments a year costs $150 per held appointment. An AI voice agent handling the after-hours share of your inbound might produce 150 held appointments annually at $40 per held appointment. These numbers are illustrative, not universal, and your actual math depends on your lead volume, source mix, and follow-up conversion rates. But the comparison illustrates why the “which is cheapest” question is the wrong question. The right question is which combination produces the most held appointments per dollar for your team’s mix.
The hidden costs nobody prices in
Salary and tools are the obvious costs. The costs that catch team leads off guard are the ones that don’t show up on the offer letter.
Ramp time. A new ISA typically takes 30 to 90 days to reach full productivity. That period costs whatever their base salary is, plus the team lead’s time coaching them, plus any missed conversions on leads they mis-handled while learning. Industry estimates suggest most team leads should budget $10,000 to $12,500 in cash reserves specifically for the ramp period.
Management overhead. ISAs need daily check-ins (15-minute morning huddles), weekly one-on-ones (30 to 60 minutes), monthly performance reviews, and periodic script and call reviews. This is real time from the team lead or a dedicated sales manager. Teams that hire an ISA and leave them alone with a script tend to see poor results and fast attrition. The management work is not optional.
Tooling stack. An ISA needs a CRM seat (kvCORE, Follow Up Boss, Sierra Interactive, BoomTown, LionDesk, Chime), a dialer (Mojo, PhoneBurner, RedX, Vulcan7 depending on inbound vs. outbound focus), a business phone line with recording, and often SMS platform tools for follow-up. Add lead source subscriptions if the ISA is running prospecting on Zillow Premier Agent, Realtor.com Connections, or paid Facebook lead campaigns. All in, tooling for a single ISA seat runs $200 to $600 per month depending on stack.
Turnover. This is where the biggest hidden cost lives. Real estate ISA turnover is high, with industry benchmarks putting annual turnover in the 60% to 80% range and average tenure at 6 to 18 months depending on team quality and role structure. Each departure costs roughly 33% of annual salary in recruiting, onboarding, and lost productivity during the gap and re-training period. For a $60,000 OTE ISA, that’s $18,000 to $22,000 per departure. Teams that go through two ISAs in a year burn nearly a full ISA’s salary just on turnover cost, before counting the leads lost during the transition.
The turnover problem is not random. It’s structural, and it deserves its own section.
When to hire an ISA (the readiness framework)
Not every team is ready for an ISA hire. In fact, most teams that struggle with ISAs are teams that hired one before the underlying business was ready to support the seat.
Four things need to be true before an ISA hire pays off:
Lead volume of 500 to 1,000 per month, minimum. Below that, the ISA doesn’t have enough phone work to justify their day. They end up either padding activity numbers or spending most of their time waiting for leads to come in. Both are demoralizing and both lead to fast turnover. If your team is generating 200 leads per month, an ISA is not the right hire yet. Grow your top of funnel first.
Cash reserves for the ramp. Budget $10,000 to $15,000 you don’t need for anything else, specifically to cover the ISA’s base pay during the 30 to 90 day ramp period. Teams that hire an ISA and then panic when they haven’t produced by day 45 tend to fire them prematurely and reinforce the turnover cycle.
Systems maturity. Your CRM needs to be clean. Your lead routing needs to work. Your scripts need to exist and be defensible. Your follow-up cadences need to be documented. Your dialer needs to be set up. If you’re hoping the ISA will fix these systems for you, you’re going to hire someone frustrated within 60 days. Systems come first, ISA comes second.
Team lead capacity to manage. This is the one most team leads underestimate. An ISA is a full-time management job on top of your existing production role. Daily huddles, weekly one-on-ones, monthly reviews, call reviews, script updates, coaching. If you don’t have the time or the temperament for that, hire a virtual ISA (where the vendor does the management) or use AI (where there’s no human to manage).
Teams that check all four boxes and hire the right person can see real ROI from an in-house ISA. Teams that check two of four and hire anyway usually contribute to the industry’s turnover statistics.
The 80% turnover problem
The ISA role has one of the highest turnover rates in real estate. Multiple industry sources put annual turnover between 60% and 80%, with average tenure often measured in months rather than years. Understanding why matters, because most of the fixes team leads try (better hiring, better training, better scripts) don’t address the underlying causes.
Repetitive work. An ISA makes 75 to 500 dials per day depending on their focus. Most of those calls end in voicemail, hang-ups, or someone who isn’t ready to buy. Even the best ISAs face constant rejection at a volume most humans can’t sustain indefinitely.
Low career ceiling. In most team structures, the ISA can either stay in the seat, get promoted into a producing agent role (which requires a license and a totally different skill set), or leave for a different company. There isn’t a clear vertical ladder for a career ISA in most brokerages. Ambitious people don’t stay in roles without visible progression.
Business hours only. The ISA works 9-to-5 or 8-to-6. Meanwhile, a large share of real estate inquiries arrive outside standard business hours — evenings and weekends are consistently the peak windows across industry lead-timing data. The ISA sees a queue in the morning that grew overnight and can only work on it during the hours when new leads are also coming in. The math is demoralizing on its face.
The 8th call problem. Industry benchmarks consistently show that 80% of eventual conversions happen between the 5th and 12th follow-up touch. But most ISAs stop after two or three, because attempts four through twelve on the same unresponsive lead feel pointless and demoralizing. The best ISAs push through this. The rest don’t, which is why teams see high call volume with modest appointment output.
The pattern that emerges: the ISA role is structurally hard because it asks a human to do repetitive rejection-heavy work at high volume during exactly the hours when the highest-intent inbound leads are unreachable. Nothing about better hiring fixes that. What fixes it is removing the parts of the job that cause the burnout, without eliminating the parts that produce value. That’s where AI enters.
Where AI fits in (and where it doesn’t)
The last 18 months have seen a rush of AI voice agent products marketed to real estate teams. Most of them are pitching one of two positions: replace your ISA entirely, or add another layer of automation on top of your existing tools. Neither framing is quite right for how AI actually fits into a real estate team’s operations.
The honest framing is that AI does some of the ISA job, dramatically better than a human can. It does not do all of it, and pretending otherwise is how vendors lose sophisticated buyers.
What AI does better than a human ISA:
- Answers inbound leads in under 10 seconds, every time, day or night. No human can hit that response time consistently. Not even a great one.
- Handles the large share of leads that arrive outside business hours. Your ISA sleeps. Your AI doesn’t.
- Absorbs volume spikes. Three inbound leads arrive during your Saturday open house. AI takes all three simultaneously. Your ISA takes one and the other two go to voicemail.
- Makes the 8th follow-up call without emotion. A human ISA gets demoralized dialing the same unresponsive lead six times. AI doesn’t care.
- Logs everything to the CRM automatically. Recordings, transcripts, qualifying answers, outcomes. No end-of-day data entry burden.
- Runs the same qualifying script perfectly every time. No skipped questions, no bad-day performance.
What a human ISA does better than AI:
- Nuanced qualification on sensitive leads (recent divorce, job loss, estate sale, first-time buyer with anxiety). AI is functional but a great human ISA is meaningfully better here.
- Handoff conversations that require reading the room. When to push for the appointment vs. when to back off. AI has some skill at this. A good human is still ahead.
- Local market knowledge that develops over months. A one-year ISA knows your inventory, your competing agents, your submarkets, your ZIP-level nuances. AI can approximate this with scripts but not replicate it.
- Coaching relationships with returning leads over time. The lead who told your ISA in January that they were still saving for a down payment and now needs to move by June. That relationship compounds. AI’s version of it is thinner.
- Cold outbound prospecting on tough lists. AI is getting better here fast but is currently weaker than a skilled human ISA on lists like FSBOs, expired listings, and pre-foreclosures.
The winning move isn’t picking between them. It’s letting each one do what it’s good at.
The hybrid model that outperforms either alone
Here’s what a hybrid ISA + AI team actually looks like in a real estate operation.
An inbound lead comes in at 9:47 PM through a Zillow Premier Agent submission. Your AI voice agent dials in five seconds, has a real conversation, confirms the buyer is looking in a specific area, gets pre-approval status, and books a Saturday 11 AM showing. The AI writes back to Follow Up Boss with the recording, transcript, qualifying answers, and appointment details. Everyone on the team wakes up to the appointment already on the calendar.
The next morning, your ISA opens the queue. The AI has already booked the after-hours leads. The queue now shows the leads that came in overnight but didn’t book (either because they hung up on the AI, said they weren’t ready yet, or needed a human touch on something the AI flagged for handoff). Your ISA works this queue with human conversations, the deeper qualifying work the AI shouldn’t be doing, and the emotional leads AI isn’t as good at.
Through the day, your ISA is on the phone with leads and doing cold outbound prospecting on your team’s chosen call list. When new inbound leads come in during business hours, the routing decides based on availability: if the ISA is on another call, AI handles it. If the ISA is free, AI still dials first (because AI is faster) and warm-transfers to the ISA in the first 30 seconds. Either way, the lead hears from a human within seconds.
At 6 PM, the ISA logs off. AI takes over. Overnight follow-up on the day’s unresponsive leads happens through AI, on cadence, on autopilot. Your ISA doesn’t have to make the demoralizing 8th call to Tuesday’s cold lead. AI does it, at 11 AM Thursday, when the lead is more likely to pick up anyway.
The result: your ISA does 6 to 8 hours of high-value work per day (which is what humans are actually good at). AI covers the other 16 to 18 hours plus the spike volume during business hours. Turnover drops because the burnout-inducing parts of the job (rejection-heavy rounds of the 8th call, dead-end voicemails at 3:47 PM, weekend inbound coverage) get handled by the AI. Your ISA stays in the seat longer because the work they’re doing is the work they’re actually good at.
Some teams report their existing ISA becoming meaningfully more productive after adding AI, not because the ISA is doing more work, but because the ISA is doing better work. They’re on the phone with warmer leads. They’re doing more of the qualification the AI hands them and less of the cold dialing that produces nothing. Their appointment set rate goes up because the leads they’re talking to are already partway through qualification when they pick up.
That is the hybrid model. It’s not AI replacing your ISA. It’s AI doing the specific parts of the ISA job that cause turnover, so your ISA can do the parts that pay.
Deciding for your team
Three common starting points for team leads reading this:
If you already have an ISA and lead volume is strong: Don’t fire your ISA. Add AI to cover after-hours, weekends, and spike volume during business hours. Your ISA becomes 30% to 50% more productive because they only handle the leads AI can’t. Your team lead time drops because AI removes the “why didn’t you follow up on Tuesday’s leads” conversations.
If you’re considering hiring your first ISA and you have $75,000 to invest: Consider hiring the ISA for 40 hours per week and adding AI for the other 128 hours per week. Split the cost. Total loaded cost lands roughly the same as one ISA alone, but coverage triples and turnover risk drops significantly because the ISA’s day is structured around what humans are actually good at, not the parts that cause burnout.
If you tried an ISA and burned out on turnover: The turnover problem is structural. Better hiring doesn’t fix it. What can fix it: rebuild the job around what humans still do well, and use AI to cover the parts that cause the burnout. Or, if you’re not ready to manage another human right now, skip the ISA hire entirely and use AI to cover the specific function (inbound response, appointment booking, follow-up cadence) that ISAs are usually hired to solve.
There is no universally right answer. There’s the answer that fits your team’s specific lead volume, systems maturity, cash position, and management capacity. But for most teams sitting between “no ISA” and “one ISA plus tools,” the hybrid model is the highest-ROI configuration available in 2026.
Frequently asked questions
What does a real estate ISA cost?
The industry standard OTE for a real estate ISA is $55,000 to $65,000 per year, split between a base salary of $30,000 to $45,000 and performance bonuses. Fully loaded first-year cost (including employer taxes, benefits, tools, ramp time, and management overhead) lands closer to $75,000 to $85,000. Virtual ISA services run $720 to $1,988 per month depending on hours. AI voice agents typically run $100 to $750 per month.
How many calls should a real estate ISA make per day?
A productive inbound-focused ISA typically makes 75 to 100 dials per day. Outbound cold prospecting ISAs (working FSBOs, expired listings, or pre-foreclosures with a power dialer) can hit 200 to 500 dials per day. Volume alone is a vanity metric. The number that matters is appointments booked from those dials.
What is a good contact rate for a real estate ISA?
Industry benchmarks put contact rate (the percentage of dials that reach a live person) at 10% to 15% for typical mixed inbound and outbound work. Appointment rate from contacts should run 20% to 30%. Show rate on booked appointments should run 50% to 70%. Below these ranges suggests either scripting or lead-quality problems worth investigating.
What is the real estate ISA turnover rate?
Industry benchmarks put annual turnover in the 60% to 80% range for real estate ISAs. Average tenure runs 6 to 18 months depending on team quality and role structure. Turnover is not primarily a hiring problem. It’s a structural feature of the role: repetitive work, constant rejection, low career ceiling, and burnout-inducing hours are all inherent to how the role is typically designed.
When should I hire a real estate ISA?
The four preconditions are: monthly lead volume of at least 500 to 1,000, cash reserves of $10,000 to $15,000 to cover the ramp period, mature systems (clean CRM, working lead routing, defensible scripts, documented follow-up cadences), and team lead capacity to manage another human. Missing any of these usually means the ISA underperforms and the team lead blames the hire.
Can AI replace a real estate ISA?
For some parts of the job, yes: inbound response inside five minutes, after-hours coverage, spike absorption, follow-up cadence on unresponsive leads, appointment booking, CRM logging. AI does these faster and cheaper than any human can. For other parts of the job (nuanced qualification, emotional handoffs, local market knowledge developed over time), a great human ISA still outperforms AI. The right question isn’t whether AI replaces the ISA. It’s which parts of the ISA workflow AI should own.
What’s the difference between a real estate ISA and a virtual assistant?
An ISA is a specialized salesperson trained in phone-based real estate conversations. Their job is to convert leads to booked appointments through phone and text. A virtual assistant is a general admin support role that handles calendars, email triage, MLS uploads, listing coordination, and other administrative work. Some virtual assistants can be trained into a light ISA function, but a real ISA is more expensive and more specialized because the sales skill set is more difficult to develop.
Should I use an in-house ISA or a virtual ISA?
In-house ISAs give you full control of scripts, brand voice, and institutional knowledge, at the cost of higher salary, ramp time, and management overhead. Virtual ISAs give you speed to production and a management layer at a lower monthly cost, but you trade away script control and consistency. For teams that can afford the fully loaded cost and have the management capacity, in-house often produces higher long-term ROI. For teams that can’t or don’t want to manage another human, virtual ISAs are the pragmatic answer. AI voice agents cover a different subset of the workflow and often work best alongside either configuration.
The bottom line
The real estate ISA role is legitimate, valuable, and increasingly hard to staff sustainably. Team leads who hire one need to know what they’re actually buying: a specialized salesperson who costs $75,000 to $85,000 all-in per year, requires ongoing management, and has an 80% chance of leaving inside two years for structural reasons that better hiring doesn’t fix.
AI voice agents in 2026 don’t replace the ISA. They replace the specific parts of the ISA job that cause the turnover, so the human ISA can focus on the specific parts that require a human. The winning real estate teams over the next few years won’t be the ones that pick between AI and their ISA. They’ll be the ones that run both, with each covering what it’s actually good at.
If you’re a team lead thinking about how to structure your team’s phone work in 2026, the hybrid model is the highest-ROI configuration available. Add AI to your existing ISA seat, or start with AI and add an ISA when your volume justifies one. Either path outperforms the traditional “hire one ISA and hope for the best” playbook that most teams are still running.
Callingly is the AI voice agent Callingly team leads use for exactly this configuration. If you’d like to see what a five-second response system looks like on your own leads, alongside your existing ISA seat, start a 14-day free trial with no credit card required.
Sources
- National Association of Realtors (2025). Home Buyers and Sellers Generational Trends Report. Available at: https://www.nar.realtor/research-and-statistics/research-reports/home-buyer-and-seller-generational-trends.
- Oldroyd, J. B., McElheran, K., & Elkington, D. (2011). “The Short Life of Online Sales Leads.” Harvard Business Review, March 2011. Available at: https://hbr.org/2011/03/the-short-life-of-online-sales-leads.
- Lead Response Management Study, James Oldroyd (via InsideSales.com), 2007. Analysis of more than 15,000 web leads and 100,000 call attempts across six companies.
- WAV Group / Weichert Real Estate (2014). Agent Responsiveness Study across 384 U.S. brokers in 11 states. Available at: https://wavgroup.com.
- ISA compensation, workflow, and turnover benchmarks referenced throughout this article draw on industry-wide reporting from Inside Sales Predictability, The Real Estate Trainer, nurtureBEAST, and Digital Maverick, alongside compensation data from ZipRecruiter and Indeed. Specific figures should be validated against your local market and lead source mix before making hiring decisions.
- Virtual ISA service pricing referenced in this article draws on publicly listed pricing from MyOutDesk, HireAiva, and comparable virtual ISA providers as of Q2 2026. Pricing changes periodically; confirm with each vendor before committing.