We Priced ISA vs AI Follow-Up at 3 Team Sizes. AI Wins Under 50 Leads.

It's 11pm and you're on Indeed, scrolling ISA job listings because your top agent just told you — again — that nobody follows up fast enough. The leads are coming in. The conversions aren't. Every coaching guru says the fix is hiring an ISA. So you start running the math, and it's more complicated than you expected: twelve hundred a month for a virtual assistant, maybe $4,500/month for someone in-house. Then you spot an AI follow-up tool for under $500/month that claims it does the same thing. Which one's actually worth it? We ran the fully loaded cost comparison at three team sizes — 30, 75, and 150 leads per month — and the standard spreadsheet turned out to be lying. Not about the close rates. It's lying about the five costs it doesn't include.

Below 50 Leads Per Month, Your First Follow-Up Hire Should Be a Bot

At fewer than 50 leads per month, AI follow-up tools cost $10–17 per lead versus $48–80 per lead for a virtual ISA at fully loaded cost. The conversion gap isn't imaginary — human ISAs do close more — but management overhead, turnover, and after-hours gaps drain that edge for small teams. Don't hire a human first.

Add a human ISA when your volume justifies the management investment. The threshold, based on our fully loaded cost analysis below, is roughly 50 leads per month — the point where an ISA reaches enough utilization to outrun the overhead.

What ISA and AI Follow-Up Actually Cost — Beyond the Sticker Price

Every ISA-vs-AI comparison you've read uses sticker prices. Tool costs $499. Virtual ISA costs $1,200. In-house ISA salary is $50,000 a year. Those numbers are real, but they're incomplete. A ShoreAgents analysis of ISA costs puts the US salary range at $50,000–$60,000 annually before benefits, while offshore virtual ISAs run $800–$1,500/month. What neither figure captures is the management time, the CRM seat, the dialer subscription, the training investment, or the cost of replacing someone who quits after eight months. We built a fully loaded cost model for all three options. Here's what it looks like when you stop lying to your spreadsheet.

The management-time line is the one that breaks most comparisons. We valued it at $100/hour — roughly what a productive agent earns when they're showing homes or running listing appointments instead of reviewing call recordings and updating ISA scripts. For a team lead already managing CRM costs, those 8–10 hours of ISA management per month come directly out of selling time. The turnover line hits even harder. According to industry data from Romano Remote, ISAs typically turn over within 8–12 months. Each replacement cycle costs a few thousand in hiring, onboarding, and lost productivity during the 2–4 week transition gap. We spread that cost across a full year in our model.

Cost Per Closing at 30, 75, and 150 Leads Per Month

Here's where the spreadsheet gets interesting. We modeled three lead-volume scenarios, and the assumptions matter — so we're showing our work. AI tools book appointments from roughly 15% of leads but produce lighter-touch qualification, resulting in a 3% lead-to-closing rate. Human ISAs don't just call — they qualify. They book from about 33% of leads with stronger pre-qualification, but they're limited to business hours. Once you factor in that 40% of web leads arrive after hours — when ISA response doesn't happen until the next morning — the ISA's blended lead-to-closing rate drops from a theoretical 8% to about 5.6%. For GCI, we've used $8,500 as the average gross commission income per closing, consistent with the per-source cost-per-close analysis we published last week.

The math at 30 leads/month tells the clearest story — and it isn't what most agents expect. The AI tool produces about 0.9 closings per month (30 leads × 3% close rate) — a cost of $554 per closing. The virtual ISA at its $2,400 true cost produces 1.68 closings (30 leads × 5.6% blended rate) — but at $1,429 per closing. The in-house ISA is worse: same 1.68 closings at $4,305 each. Yes, the human ISA generates more total closings. But the cost-per-closing gap means you'd need to be confident the ISA's extra 0.78 closings per month — worth $6,630 in GCI — will actually materialize. And at 30 leads per month, that confidence is misplaced. Here's why.

Five ISA Costs the Standard Spreadsheet Leaves Out

The conversion numbers above assume your ISA performs at benchmark from day one and never quits. In practice, five costs erode the ISA's math — especially at low lead volumes where there's no margin for error.

1. Management time eats your selling hours. Plan on 8–12 hours per month: weekly one-on-ones, call reviews, script updates, performance tracking. For a solo agent producing $100/hour in GCI, that's $800–$1,200/month in opportunity cost. At 30 leads, you're spending 8 hours managing someone who handles 1.5 leads per working day. A team lead running six agents has the bandwidth. A solo agent doesn't. From what I've seen across the teams we track, first-time ISA managers consistently underestimate this by 40–60%. They budget four hours. They spend ten.

2. After-hours leads get stale. Nearly four out of ten web leads arrive between 6pm and 8am or on weekends, according to lead-routing data from platforms like the AI tools we recently compared. Your ISA isn't working at 10pm. The lead that hit your CRM at 9:47pm gets a callback at 9:15am — twelve hours later. Industry data from the Structurely team and others consistently shows that leads contacted within five minutes are 21 times more likely to qualify than those contacted after 30 minutes. An AI tool responds in seconds, around the clock. That's what closes the conversion gap at low volumes — not better scripts, not smarter routing, just being there when the lead is hot.

3. Turnover wrecks small operations. The average ISA tenure runs 8–12 months. Every departure doesn't just cost money — it triggers a 2–4 week gap where leads get minimal follow-up during hiring and onboarding. For a team doing 75+ leads/month, that gap isn't ideal but it's survivable. For a solo agent at 30 leads, one turnover event means a dozen leads with no human follow-up. At the blended close rate from our model, that's roughly half a lost closing — over four thousand in GCI, gone. We amortized those replacement costs in the table above, but the real damage isn't the hiring line item — it's the conversion disruption during the gap.

4. Hiring overhead adds up. Job postings, screening calls, interviews, reference checks, and onboarding documentation — it doesn't sound expensive until you're doing it for the second time in nine months. For virtual ISA services, the provider handles recruiting, so this cost is minimal. For in-house hires, budget two to three thousand per replacement in hard costs and 15–20 hours of your time. Most teams with a single ISA go through this cycle at least once per year. That's time you won't spend prospecting or closing.

5. Ramp-up creates a dead zone. A new ISA doesn't convert at benchmark rates from week one. The ramp-up to full productivity takes 2–4 weeks for experienced hires and 6–8 weeks for someone new to real estate ISA work. During that ramp, you're paying full cost for partial output. At the virtual ISA's true cost, a 3-week ramp at 50% productivity wastes roughly $900. Not catastrophic on its own, but combined with the turnover cycle, you might lose 6–8 weeks of full performance per year.

How to Build Your Follow-Up Stack by Lead Volume

Here's my honest take: the ISA-vs-AI question is the wrong framing. It's not either/or. It's a sequencing decision. And the sequence depends entirely on how many leads you're running per month. I've watched teams blow five grand a month on an in-house ISA when they're generating 25 leads. I've also seen AI-only shops leave money on the table at 100+ leads because nobody's having the nuanced qualification conversations that close deals. The right answer changes with volume. Here's the same kind of team-size math we applied to transaction management.

The hybrid approach at 100+ leads works because each tool covers the other's weakness. AI doesn't miss a single lead — it catches the after-hours inquiries that'd otherwise go cold — and handles the initial qualification conversation. When a lead signals genuine intent (asks about pricing, requests a showing, mentions a timeline), the AI hands off to the human ISA for the appointment-setting conversation that requires judgment and rapport. One team lead we spoke with described it as "the bot fishes, the ISA nets." It's a clean division: AI catches leads the ISA would miss, ISA converts leads the AI can't close. Combined, you're getting higher contact rates AND higher qualification quality.

Here's what I'd tell any team lead running this decision right now: don't hire a human ISA as your first follow-up investment. Deploy an AI tool at the price point we showed above. Let it run for 60–90 days. Track your contact rate, appointment rate, and cost per closing. Then — and only then — layer in a virtual ISA to handle the leads that need a human touch. You'll have baseline data to measure the ISA's actual impact, and you won't be scrambling if the ISA doesn't work out.

Frequently Asked Questions: ISA vs AI Follow-Up for Real Estate Teams

What's the minimum lead volume to justify hiring an ISA?

Based on our cost analysis, 50 leads per month is the threshold where a virtual ISA's higher conversion rate consistently outweighs the management overhead. Below that, the ISA doesn't get enough volume to justify the time you'll spend managing it — those 8–12 hours per month come straight out of your selling schedule. At 50+ leads, the ISA reaches roughly 50% utilization and generates enough extra closings to justify the investment several times over.

Can AI follow-up tools actually set appointments?

Yes. Tools like Structurely ($299–$499/month) and Ylopo's rAIya ($795/month bundled) handle initial lead conversations via text and chat, qualify based on intent signals, and book appointments directly into your CRM calendar. They won't match a skilled human ISA's qualification depth, but they respond instantly at 2am — which matters more than qualification nuance for low-volume teams.

Does an AI tool replace the need for a CRM?

No. AI follow-up tools integrate with your existing CRM — Follow Up Boss, kvCORE, Sierra Interactive, or whatever you're running. The AI handles the conversation layer; your CRM's still handling the data, pipeline management, and long-term nurture sequences. Think of it as a module that sits on top of your CRM, not a replacement. It won't track your deals or manage your database — that's still your CRM's job.

What about ISA services that provide the person for you?

Virtual ISA services like MyOutDesk and ShoreAgents handle hiring and basic training, which eliminates some overhead. But you still manage the daily workflow, review calls, update scripts, and handle performance issues. The management burden drops from 10–12 hours to 6–8 hours per month — better, but not zero. And turnover still happens; you just don't pay the recruiting cost.

Ready to See How RobinFlow Handles Team Follow-Up?

Whether you're deploying AI, managing an ISA, or running both, the stack only works if your CRM keeps pace. See how robinflow's team features handle lead routing, follow-up tracking, and team accountability — built for the exact workflows we covered above.