How a Real Estate VA Handles Showing Scheduling
It is 4:12 p.m. on a Thursday and an agent is sitting in her car outside a house in Linden with the engine running, because the buyers she has been waiting twenty minutes for are not coming. They told her at 3:47 p.m. on WhatsApp. She did not see it, because at 3:47 p.m. she was inside a different house in Emmarentia explaining to a couple why the servitude on the title deed is not the problem they think it is.
Three other things are happening while she sits there.
A Property24 enquiry from 11:04 that morning is still sitting unread — a buyer asking whether the Parkhurst listing is available to view this weekend. He has, by now, sent the identical enquiry to two other agents. One of them replied at 11:09.
The Parkhurst seller has phoned twice since Tuesday to ask why nobody is viewing. Two people did view, on Saturday. One said the road noise from Jan Smuts was a problem; the other said the kitchen looked smaller than the photographs suggested. Neither comment has reached the seller, because the feedback lives in a WhatsApp thread nobody has had a free half-hour to read back through.
And on Saturday she has three properties on show, two of them booked for 2 p.m. to 4 p.m. She booked one on Tuesday and one on Wednesday and did not notice, because the diary is a Google Calendar, a WhatsApp thread, a portal notification and — for two sellers who refuse to use anything else — a series of phone calls.
None of this is a discipline problem. She is not disorganised. She is a competent professional performing a coordination function that, in most agencies, has not been assigned to anybody — it has simply been absorbed, in the gaps, by the person in the business whose hour is worth the most.
The Four-Message Problem
Start with the anatomy of a single showing request, because the arithmetic is where this stops being a feeling and becomes a staffing question.
A 2026 workflow analysis of agent scheduling puts the typical showing request at four to six messages: a buyer asks about a property, the agent checks the diary, proposes times, the buyer cannot make those work, more back-and-forth follows, and eventually something is confirmed. That is one property, one buyer, in the best case where nobody’s plans change. The same analysis found agents managing more than five to ten active buyer relationships spending three to five hours a week on scheduling coordination, with fifteen active buyers in a busy season compounding that friction into eight to twelve hours a month of pure coordination — work that does not help anybody find or sell a home.
That figure is the optimistic version. It counts the messages. It does not count the double bookings, the absence of travel buffers between properties on opposite sides of a city, the no-shows that happen because six text messages produced a confirmation nobody wrote down, or the interested buyer who quietly found an agent who replied faster.
One scheduling vendor put the structural point better than most industry commentary manages: a real estate agent’s calendar is not really a calendar, it is a coordination problem with five other humans on each row — the buyer, the listing agent, the seller, the lender, and the inspector or appraiser once the deal moves. Add a weekend of show days, two listing presentations and a transfer that needs chasing, and the week breaks any tool designed for someone booking half-hour video calls.
The load lands on people already at capacity. Industry estimates put roughly 36% of an agent’s professional time into administrative work, and NAR’s 2025 agent productivity survey puts showing coordination specifically — calls, confirmations, feedback collection and reschedules — at an average of two and a half hours a week. Both numbers sit inside a job where weekends are prime time and a Saturday can easily mean five or six hours driving clients between suburbs.
A showing request is four to six messages. Fifteen active buyers turns that into eight to twelve hours a month of coordination that produces nothing except the appointment itself.
Why the Diary Became Part of the Product
There is a temptation to file all of this under admin — necessary, tedious, secondary to the real work of selling. That framing was defensible when buyers were patient. It is not defensible now, because the ease of getting into a property has become one of the things buyers actively compare.
Rently’s 2026 Renter Touring Expectations Report makes the point about rentals, but it generalises: search has become a phone-first process where people compare listings, prices, neighbourhoods and photographs before they leave the sofa — and then, when they are ready to see a home in person, the process becomes manual and slow again. Call, email, wait, repeat the same information, work around office hours. One renter in that research described driving an hour and a half to a property, finding the access instructions wrong, being unable to reach anyone on the support number, and going home again.
The volume data shows why this matters more than it used to. ShowMojo’s Q1 2026 update — drawn from over 78 million leads, 25 million showings and 705 million emails, texts and calls — found that 61% of rental enquiries now arrive outside standard business hours, and that the average listing generated just 35.4 leads and 18.1 scheduled showings, the lowest first-quarter lead volume on record. Fewer enquiries per listing, and nearly two thirds of them landing when the desk is empty.
Sales behaves the same way. The average agent takes 917 minutes to respond to a new lead, against long-standing research from MIT and InsideSales.com showing a five-minute response makes qualification twenty-one times more likely. In South Africa that gap has a specific shape: buyers browse Property24, Private Property and Gumtree simultaneously and send the same enquiry to several agents at once, so the agent who replies first usually gets the viewing — and most reply in hours, because they are in traffic, or standing in a property in Fourways when the enquiry lands on a Midrand listing.
Meanwhile the underlying market is not soft. FNB’s estate agent activity rating sat at 6.3 out of 10 in Q1 2026, above its long-term average. National time on market dropped to 11.6 weeks against a long-term average of thirteen — six weeks in the Western Cape, a sluggish fifteen in Gauteng. First-time buyers made up about 40% of the market, the highest share since FNB began tracking in 2004. There are buyers, and there is competition for them. The first competitive event in the sequence is not the offer or the negotiation. It is whether somebody answered and got a diary entry confirmed.
What Showing Coordination Actually Involves
Most agents underestimate this work because they experience it in fragments. Written out end to end, it is a nine-part job.
Enquiry triage and first response. Portal leads, WhatsApp messages, calls, website forms and the board outside the house all produce enquiries in different places. Somebody has to consolidate them, respond, and do it quickly enough to matter.
Qualification before the calendar invite. Are they pre-approved? Are they selling first? Have they seen the price? Nothing is more expensive than a Saturday spent showing a R4.5 million house to somebody approved for R2.8 million. Every showing is a real investment of the listing agent’s time and a security consideration for the homeowner, and disciplined agents qualify rather than simply opening doors.
Access logistics. The part that never appears in software demonstrations and eats the most time in practice. Whether the seller needs to be out. Whether there is a tenant legally entitled to notice. Whether the complex needs a name at the gate an hour ahead. Whether there is a dog. Whether the key is with the neighbour, in a lockbox, or with a domestic worker who finishes at two.
The booking and its geography. Three viewings in one afternoon is efficient only if they are sequenced sensibly. Booked in the order the requests arrived, they are a driving itinerary designed by nobody.
Confirmation and the reminder cadence. Covered below, because this is where most of the recoverable value sits.
Reschedules and cancellations. Which is to say: the actual job. Nothing about a viewing diary is stable.
Post-showing feedback collection. Manually chasing feedback recovers responses on only 30–40% of showings, while structured surveys average 60–75%. Even experienced agencies acknowledge typical capture of about 40% to 45%.
Seller reporting. The seller does not want a spreadsheet. They want to know who came, what they said, and what it means — weekly, without asking twice.
CRM hygiene and the follow-up trigger. The agents growing fastest treat every showing as the start of a follow-up sequence rather than a one-off event. That only works if somebody logs the showing.
Now read that list again and ask which items require a registered property practitioner exercising professional judgement. Exactly one: being in the room. The valuation conversation, the negotiation, the reading of a buyer’s face when they walk into the main bedroom — those are the agent’s job and cannot be given away. The other eight items are coordination, and coordination is a role, not a personality trait.
The only part of showing coordination that requires a registered property practitioner is the showing itself. Everything around it is a scheduling job that most agencies have never actually assigned to anyone.
Why Capable Agents Still Lose Viewings
The reason this work degrades is structural, and worth being precise about, because agents tend to blame themselves for it.
Showing coordination is continuous work with soft deadlines. Selling is episodic work with hard deadlines and a person on the other end. When the two compete, the episodic work wins every time — not because the agent is undisciplined, but because that is what deadlines do.
The interruption research makes it worse. ActivTrak puts the average focused work session at roughly thirteen minutes, down about 9% on 2023, and Gloria Mark’s research at UC Irvine found it takes over twenty-three minutes to return to a task after a single interruption. Now apply that to someone who is not at a desk at all — driving between suburbs, standing in a stranger’s kitchen, on a call with a bond originator. Coordination that depends on knowing where you left off cannot be performed in the gaps between other people’s appointments.
And the demand curve runs against the agent. Those enquiries arriving outside business hours arrive precisely when she is at a show day, at a valuation, or asleep. The evening enquiry gets answered on Monday. The buyer books with someone else on Friday.
The No-Show Is a Design Problem
Here is where the argument stops being about time saved and starts being about outcomes, because there is a genuinely large body of evidence on missed appointments — most of it from medicine, all of it directly applicable.
A systematic review of the reminder literature found 95% of studies reporting a positive effect on attendance, with an average 41% reduction in missed appointments. A separate review reported a weighted mean relative reduction in non-attendance of 34% from baseline, with 28 of 29 studies showing improvement.
But the more interesting findings are about how the reminder is delivered, not whether one is sent.
A quality-improvement study across 250 primary care patients found live reminders produced a 3% no-show rate; messages left as voicemail produced 24%; calls that went unanswered produced 39%. The same intervention, three times, with an eightfold spread depending on whether a human being actually reached the person.
Content matters too. Two randomised controlled trials across roughly 20,000 UK outpatients found that rephrasing reminders cut missed appointments by around a quarter in relative terms — a difference between best and worst wording comparable in size to the entire effect of sending reminders at all. A larger Israeli trial randomised 161,587 Clalit Health Services members across fourteen hospitals into nine reminder groups; the best-performing frame cut no-shows from 21.1% to 14.2% and lifted advance cancellations from 17.2% to 26.3%.
That second number is the one property people should care about most. Advance cancellation is not a failure. It is a Saturday slot you can resell to another buyer instead of standing in a driveway in Linden with the engine running.
Property vendors report figures in the same direction — one self-showing platform claims a 75% reduction in in-person no-shows — but that is vendor data on a self-selecting customer base, and the peer-reviewed work is the sturdier ground. What it says is consistent: reminders work, two-way beats one-way, live human contact beats an unanswered dial tone by a wide margin, and the wording is not a detail.
None of which is difficult. All of which requires somebody to have the time.
Live reminders produced a 3% no-show rate. Voicemails produced 24%. Unanswered calls produced 39%. The intervention is identical — what changes is whether a person actually made contact.
The Human in the Loop
The obvious objection is that this is exactly what software is for, and in 2026 that objection increasingly means a conversational AI agent answering the phone and booking the viewing.
Some of that is genuinely good. Automated dispatch of reminder sequences is better than a person remembering. Calendar conflict detection is better than a person noticing. Transcription, first-draft feedback summaries, availability lookups, routine confirmations — all of it is real, and any agency not using it is leaving time on the table.
The trouble starts when the automation is the whole system rather than a layer inside one.
Gartner’s 2026 research attributes 57% of failed AI initiatives to unrealistic expectations and 38% to poor data quality. More specifically for voice deployments, five failure modes account for most production breakdowns — speech hallucination, persona drift, accent and dialect recognition bias, security threats, and escalation failure — and none of them are reliably visible in a well-managed pilot. Speech hallucination is when the system makes commitments nobody authorised, and without retrieval grounding the hallucination-related complaint rate runs at roughly 0.34% in production — about 340 customer-visible incidents per 100,000 monthly calls.
The transcription layer beneath these systems has its own failure profile. Research on OpenAI’s Whisper found roughly 1% of audio transcriptions contained entire hallucinated phrases absent from the underlying audio, with 38% of those including explicit harms such as inaccurate associations or implied false authority.
Two of those failure modes deserve particular attention in South African property. The first is accent and dialect recognition bias: an agency in Johannesburg fields calls from buyers whose first language is isiZulu, Sesotho, Afrikaans, Portuguese or Mandarin, on cellular connections, often from a car. A system that mishears a surname or a suburb does not announce that it has done so.
The second matters most for scheduling specifically. As one 2026 analysis of voice AI monitoring puts it, the danger is silent failures, where the conversation sounds successful but the actions do not complete. Everyone hangs up satisfied. Nothing gets into the diary. Nobody finds out until a buyer is standing at a gate in Bryanston at 10 a.m. on a Saturday and the seller is in Ballito.
Related, and just as expensive: quoting a property that was let or sold yesterday is the single most common cause of a bad first impression in lettings, because the availability data behind the system is static. And operations teams that deployed scheduling agents have found them looping indefinitely when a caller deviates from the expected script — which callers do, constantly, because a real enquiry sounds like “hi, I saw the one in Greenside, is it still the R3.4 or did that change, and can my mother come too, she’s the one actually buying.”
There is a legal edge as well. Practitioner analysis of voice-agent deployments in 2026 notes that courts increasingly hold companies to the promises made by their chatbots, and that regulators do not accept “the AI hallucinated” as a defence. A system that confirms a viewing the seller never agreed to has not made a scheduling error. It has made a commitment on a property practitioner’s behalf.
Then there is the part of showing coordination that is not a customer-service function at all. It is a safety function. Figures cited by agent-safety platform Tether RE, selected for NAR’s 2025 REACH programme, put 600,000 REALTORS® meeting a client alone to show a property in 2024, with more than 300,000 reporting they felt unsafe. NAR’s broker guidance is process rather than paranoia: a written policy requiring identification from prospective clients before they are shown a property, and a copy of ID from everyone over eighteen attending, vetted individually rather than only for whoever made the booking. Somebody has to collect that, before the appointment, in a way that does not feel like an interrogation.
Collecting it also drops the agency straight into two South African statutes. Estate agents are accountable institutions under Schedule 1 of the FIC Act, and POPIA carries penalties of up to R10 million or ten years’ imprisonment for serious infringements involving the unlawful processing of personal information. Customer due diligence is triggered when the business relationship is created — before marketing starts — not at transfer. A showing register full of ID numbers, contact details and financial indications is exactly the kind of file that a regulator asks pointed questions about, and exactly the kind of file that gets kept in a WhatsApp thread on a personal phone.
So the argument here is not that AI is bad at this work. It is that the work contains a continuous stream of judgement calls — whether a caller who will not give a surname should be given an address, whether “we’ll come Saturday” from a buyer who has cancelled twice needs a confirming call, whether a seller’s “any time is fine” survives an actual 7 a.m. request, whether the enquiry that just came in is a serious buyer or the neighbour — and judgement is precisely the thing that does not survive being handed to a script. The right architecture is a trained person operating good software, which is also, not coincidentally, where the whole industry has landed: enterprises are moving away from fully autonomous bots toward an AI-assisted, human-led approach.
The South African Advantage
If the conclusion is that showing coordination needs a person, the next question is where that person should sit. There is a specific case for South Africa, and it is not primarily about money.
The clock is the whole argument
South Africa runs on GMT+2 with no daylight saving shift, so the relationship with the UK is a fixed one to two hours rather than something that moves twice a year. A South African VA sits inside the entire European working day and catches the US East Coast morning with three to five hours of genuine overlap.
For a South African agency the case is simpler still: it is the same hour. An enquiry that lands at 11:04 on a Thursday is handled at 11:06, not overnight. For an international agency the shape is a shift extension — the agent finishes her last viewing at 5 p.m.; the VA has the evening enquiries triaged, the Saturday diary sequenced by geography, and the confirmations out before she opens her laptop the next morning.
Set that against the alternative most often proposed. The Philippines sits at GMT+8 — seven to eight hours ahead of the UK, six ahead of South Africa. In a workflow where most enquiries arrive outside business hours and the competitive event is measured in minutes, a round trip that costs a day is not a minor inefficiency. It is the difference between booking the viewing and reading about it.
English, register, and the fact that this job happens on the phone
South Africa scores 602 on the EF English Proficiency Index, ranking 13th of 113 countries globally, first in Africa and ahead of both the Philippines and India. Accent neutrality is one of the drivers behind South Africa’s rising position in buyer-preference tracking of offshore service markets.
But proficiency is table stakes. Register is the real point, because showing coordination is composed almost entirely of delicate conversations. Telling a seller nobody came this week. Telling a buyer the Saturday slot has gone. Asking a stranger for a copy of their ID without implying you think they are a criminal. Chasing a buyer’s agent for feedback for the third time without souring a relationship you will need again next month. South African professional English sits between British restraint and American directness, which happens to be the exact register those conversations require.
Property fluency, not general office experience
The difference between a competent administrator and a useful one shows up in the details of access. Knowing that a sectional title complex may need the trustees or the managing agent notified. Knowing that a tenant in occupation has rights and a landlord’s convenience does not override them. Knowing that “the gate code” and “the estate’s visitor protocol” are two different things and both can fail on a Saturday morning.
VAConnect recruits for property roles on the basis of prior property management, conveyancing or estate agency experience rather than general administrative background, which is why its real estate VAs arrive already fluent in the vocabulary. The measured quality picture supports the pattern more broadly: BPESA and Invest SA’s GBS Investor Handbook reports roughly 18% higher customer satisfaction from South African providers than Indian and Philippine comparables, translating to 4–5% better client retention year on year.
Cost, deliberately last
South African professional support runs 40–60% below UK, US and Australian equivalents, and VAConnect’s managed placements start from $1,088 a month. Set that against a local listings-and-viewings administrator plus UIF, a desk, a recruitment fee, and the three months of nothing that follow a resignation in September.
But cost is the least interesting part of the argument, and any provider leading with it is telling you what they compete on.
The number that actually decides this is attrition. South Africa runs at 10–18% against 30–40% in the Philippines and 30–35% in India. That is decisive for showing coordination specifically, because a scheduling VA’s value is almost entirely accumulated context — which seller insists on twenty-four hours’ notice, which complex needs a name at the gate the day before, which buyer’s agent always runs fifteen minutes late, which tenant will not answer the phone but replies to WhatsApp within a minute, which two suburbs should never be booked in the same afternoon because of the N1. None of that lives in a CRM field. All of it walks out the door when the person does.
Every eighteen months, a Philippines-based coordinator resets to zero on everything they have learned about your sellers, your complexes and your traffic. A South African one, statistically, does not.
Managed, Not Matched
The matching model — a marketplace, a shortlist, a contractor you engage directly — fails at this work in three predictable ways.
The training cost lands entirely on the agency, and resets. You spend six weeks teaching someone your access protocols, your seller reporting format and your qualification thresholds. That investment is yours, uninsured, and when they leave for a better-paying client you start again.
The quality is unverified until it fails publicly. Nobody discovers a scheduling problem quietly. It is discovered when a seller phones the principal, or a buyer’s agent posts about being left standing at a gate, or a Saturday’s worth of viewings turns out to have been double-booked against a show day.
And there is no continuity. Viewings do not pause for illness. A missed Saturday is not a delayed task — it is a lost weekend of buyers in a market where the average listing generates fewer enquiries than it did last year.
VAConnect’s model is built against exactly those three failure modes. VAs are employees of the agency rather than contractors dividing attention between a dozen clients. Sourcing runs through VAJobs.co.za with skills testing, background checks and cultural-fit assessment before anyone reaches a shortlist; training runs through VAVarsity before a VA touches a client system; Atomic Energy monitors wellbeing and workload and VAPIness provides accountability in both directions. An account manager owns the quality outcome, with monthly performance reviews. If the match is not working, the replacement is free and the transition managed — no fees, no friction — so the onboarding investment survives. In seventeen years that has been necessary fewer than eight times.
The published REVA service page covers precisely this territory: scheduling, confirmation emails, feedback collection, sign and lockbox coordination, and post-showing follow-up, alongside listing and transaction support. Every package includes managed quality, backup cover, SOP creation for handover and priority stand-in availability.
The founder’s framing has been consistent since she built the model in 2014, six years after starting the business as Lime Tree Consulting: “I don’t want to be the biggest VA company. I want to be the one where nobody leaves — not the clients, and not the VAs.” The 98% client retention figure is a consequence of that design, not an accident of it.
One verified client review from the property side puts the effect in the terms agents actually feel: “The REVA handles everything between offer and settlement — documents, comms, scheduling. My close rate improved because I was focused on clients, not admin.”
The First Ninety Days
The failure mode when agents delegate scheduling is handing over everything at once and then spending more time supervising than they saved.
Weeks one and two: the boring work first. Confirmations and reminders only — no qualification, no seller conversations. The confirmation when a viewing is booked, the reminder the day before, the two-way check on the morning of. Deliberately low-stakes work with an immediately measurable output, which is where calibration happens.
Weeks three to six: build the artefacts. Three documents, written by the VA and approved by the agent. An access brief per active listing — who has the key, who must be notified, the gate protocol, whether there is a tenant, whether there is a dog, what notice applies. A qualification script covering the questions that must be asked before a slot is confirmed, and the ones that must not. A feedback template short enough that a busy buyer’s agent will actually complete it. Then the reminder cadence gets designed properly, with wording the research says is worth as much as the reminder itself.
Weeks six to twelve: extend and measure. The VA takes over enquiry triage, geographic sequencing of the diary, feedback chasing and the weekly seller report. Three numbers get tracked from week six: median time from enquiry to confirmed viewing, no-show rate, and feedback capture against the 40–45% baseline most agencies run.
The day-ninety test is simple. Can you state your no-show rate and your median enquiry-to-viewing time from memory? If you can, the function exists. If you cannot, it is still being absorbed in the gaps by whoever happens to be free.
The Gap Is Wider Than It Looks
What is startling, reading the 2026 data together, is not that coordination is hard. It is how large the difference has become between agencies that treat showing scheduling as a role and agencies that treat it as something the agent does in the car.
One group answers the 11:04 enquiry at 11:06 and books the Saturday slot; the other answers on Monday and finds it taken. One runs a two-way reminder cadence and loses a handful of appointments a quarter; the other loses one in five and calls it the market. One walks into a price conversation with feedback from three quarters of its viewings; the other has 40% capture and an opinion. One seller gets a Friday report without asking; the other phones twice and starts wondering about the mandate.
Each difference is small in isolation. Compounded across a listing book over eleven and a half weeks on market, they are the difference between an agency that grows and one that stays exactly the size of its principal’s diary.
The buyers are there. First-time buyers are at their highest share since 2004 and the activity rating is above its long-term average. The constraint is not demand. It is whether anybody had a free ten minutes to reply.
DIY Coordination vs Generic Freelancer or AI Tool vs VAConnect Managed Real Estate VA
| DIY / Agent-Coordinated | Generic Freelancer or AI Scheduling Tool | VAConnect Managed Real Estate VA | |
|---|---|---|---|
| Enquiry response time | Hours to days; depends on whether the agent is in a viewing | Seconds for automated reply, but no qualification behind it | Minutes, inside the same working hour, with qualification attached |
| After-hours enquiries (61% of volume) | Answered next business day | Auto-acknowledged; nothing enters the diary reliably | Triaged same evening or first thing, with overlap on UK and EU hours |
| Qualification before booking | Inconsistent; skipped when busy | Not performed, or scripted and brittle | Scripted, applied consistently, escalated when unclear |
| Access logistics (tenants, gates, keys, dogs) | Held in the agent’s head | Not modelled; static data goes stale | Documented per listing in an access brief, kept current |
| Diary sequencing by geography | Booked in the order requests arrive | Slot-based; no travel buffer logic | Sequenced deliberately to cut drive time |
| Double bookings | Common across calendar, WhatsApp and portal | Reduced within the tool; fails across tools | Single owner of the diary across all channels |
| Reminder cadence | Ad hoc, often skipped | One-way automated messages | Multi-touch, two-way confirmation, wording built on the reminder research |
| No-show handling | Discovered on arrival | Logged after the fact | Confirmed in advance; cancellations converted to reopened slots |
| Feedback capture | 30–40% via manual chasing | Survey sent; low completion, no follow-up | Chased persistently to a target well above the 40–45% norm |
| Seller reporting | When the seller asks, usually twice | Raw data export | Weekly written report with interpretation, unprompted |
| Prospect ID and safety vetting | Intended, rarely enforced | Not collected | Collected before the appointment as standard process |
| POPIA / FICA handling of showing data | Personal phone, WhatsApp threads | Third-party platform, unclear processing terms | Structured records, NDA and data-protection frameworks in place |
| Judgement calls (odd requests, evasive callers) | Made well, when the agent has time | Loops, escalates poorly, or invents an answer | Made by a trained person who escalates when uncertain |
| CRM logging and follow-up trigger | Retrospective, patchy | Logged only if the tool owns the workflow | Logged same day; follow-up sequence triggered |
| Continuity during illness or leave | Agency stops | None | Backup cover and priority stand-in availability included |
| Accumulated context (sellers, complexes, traffic) | Held by the agent | Resets at every engagement | Compounds; retained at 10–18% attrition vs 30–40% offshore norms |
| Training investment | Borne by the agency, lost on exit | Borne by the agency, resets | Held by VAConnect; VAVarsity training before client systems |
| Quality accountability | The agent’s, on top of everything else | The agency’s, discovered on failure | Account manager, monthly reviews, free replacement — no fees, no friction |
| Cost | Free, paid in the agent’s highest-value hours | Cheapest per hour; most expensive per error | From $1,088/month, managed, 40–60% below onshore equivalents |
Ready to stop coordinating from the driver’s seat? Book a 30-minute discovery call and we’ll map your current enquiry-to-viewing workflow and show you exactly where the appointments are leaking. → Real Estate VA — vaconnect.co.za
Sources
Academic and peer-reviewed
- Systematic review of patient reminders and missed appointments, Pan African Medical Journal / One Health — 95% of studies showing positive effect, average 41% reduction in missed appointments
- Hallsworth, Berry, Sanders et al., “Stating Appointment Costs in SMS Reminders Reduces Missed Hospital Appointments: Findings from Two Randomised Controlled Trials,” PLOS One — ~20,000 participants across two UK trials
- A/B testing of pre-appointment SMS framing, Clalit Health Services, 161,587 members across 14 hospitals, nine message conditions
- “No-Show Rates When Phone Appointment Reminders Are Not Directly Delivered,” Psychiatric Services — live (3%) vs voicemail (24%) vs unanswered (39%)
- Gloria Mark, University of California, Irvine — interruption recovery research
Industry
- ShowMojo Q1 2026 Rental Market Update — 78M leads, 25M showings, 705M communications
- Rently 2026 Renter Touring Expectations Report
- NAR 2025 agent productivity survey; NAR REALTOR® Safety resources and Prospect Identification Form guidance
- FNB Estate Agent Survey / RODE Property Report Q1 2026
- Lightstone residential transaction and agent sentiment data
- Gartner 2026 AI initiative failure research; voice AI production failure-mode analysis
- BPESA / Invest SA GBS Investor Handbook; EF English Proficiency Index
Regulatory (South Africa)
- Financial Intelligence Centre Act, Schedule 1 — property practitioners as accountable institutions
- Protection of Personal Information Act 4 of 2013 — penalty provisions
- Property Practitioners Regulatory Authority (PPRA) — Fidelity Fund Certificate requirements
Forum, consumer and practitioner sentiment
- Renter verbatim accounts of failed access instructions and wasted trips (Rently 2026 report)
- South African estate agent lead-response commentary on simultaneous Property24 / Private Property / Gumtree enquiries
VAConnect
- vaconnect.co.za — Real Estate VA (REVA) service scope, pricing, replacement guarantee, retention data
- Karen van Zyl, founder — on retention as a design goal
- Verified Clutch client reviews