It is 2:10 pm. You have locked flat 302, dropped the key at reception, and the next viewing is at three, on the other side of the main road. Four unplayed voice notes on your phone: the couple looking at the two-bed sent 4 min 12 s, the owner of the corner house sent two of just over a minute each, and the buyer who asked for the floor plan sent 38 seconds.
You put the earphones in and play the first one while driving. Halfway through comes the line that matters: she will go to 620, but the kitchen units stay. You write nothing down. You are driving.
At seven in the evening, at home, you open the CRM to log the offer and hesitate. Was it 620 or 630? Back into the voice note. The four minutes again, this time hunting for the exact second she said the number.
This piece is about that part of the day: the stretch between the voice note arriving and the record that stays.
The channel moved to voice. The record stayed written
Conversations with buyers and with owners moved to the voice message, and not by a little. In 2022, WhatsApp reported an average of 7 billion voice messages a day on the platform. On the business side, the ICT Enterprises 2024 survey, run by Cetic.br in Brazil, shows WhatsApp or Telegram as the leading online platforms used by Brazilian companies to interact with customers: 74% of them in 2024, against 42% in 2017. The same survey records that 49% of the companies that sold over the Internet in the previous twelve months closed through WhatsApp, Skype or Facebook chat messages, down from 55% in 2023.
And there are plenty of people absorbing that volume. COFECI, the federal council for the profession in Brazil, records more than 730,000 registered estate agents in the country. Most work with a phone, a car and a diary — not with a contact centre behind them.
What did not move is the record. The offer goes into the CRM in writing. The owner's message goes into an email in writing. The condition the buyer accepted goes into the team chat in writing. Between the voice note and the text, someone listens and types. That someone is you, at seven in the evening.
The three kinds of voice note an agent gets
They are not all the same, and separating them helps you decide what to do with each.
The first is the short operational one: confirming a time, asking the concierge's name, warning you about a twenty-minute delay. Under thirty seconds, needing a quick reply rather than a record.
The second is the owner explaining the whole property in one go — floor area, the 2019 refurbishment, the service charge, what stays and what goes. It runs well past three minutes, and it is the one most worth keeping as text, because you will repeat it to every prospective viewer.
The third is the negotiation one: figure, condition, deadline, and the caveat added at the end, when the note seemed to be over. Short, decisive, and the only one you will reopen three times if it is not written down somewhere.
The maths of your day
Worth doing once. The sum below is illustrative — swap the numbers for yours.
Say 22 voice notes arrive from buyers and owners in a day, averaging 1 min 40 s. That is 36 minutes of listening alone. Playing at 1.5x brings it down to 24. Now add the rest: finding the voice note in the chat, scrubbing back to the figure, opening the CRM, typing the two lines that have to stay. At 40 seconds per note, that is another 15 minutes. Close to 40 minutes a day — and, across five days, three hours stuck in a stretch that is not a viewing, not a listing and not a negotiation.
It may be less in your case. In launch week it is usually a good deal more. The point of the sum is not the total: it is noticing that the stretch exists and that it shows up in no report anywhere.
What exists today, and where each option stops
Five options. None of them is wrong — each one stops somewhere different.
| Option | What it fixes | Where it stops |
|---|---|---|
| Playing at 1.5x or 2x | Cuts roughly a third off listening time | You still listen to everything, and still type up what has to be recorded |
| WhatsApp native transcription, on the phone | Gives the voice note text with nothing installed | WhatsApp announced the feature for the phone, generated on the device itself and triggered by a long press: one note at a time, away from the computer screen where the CRM is open |
| Asking the client to type instead | Works when they agree | An owner in a hurry and a buyer behind the wheel send a voice note again in the next message |
| Forwarding the voice note to the team | Takes the listening off your hands | Hands the work to someone else and spreads the client's details across one more device |
| Transcribing on the computer, inside WhatsApp Web | The voice note arrives as text on the same screen where you log it, ready to copy | Needs the computer on with the tab open, and respects daily quotas by plan |
How this works in ZapVox, step by step
ZapVox is an extension for WhatsApp Web — the tab already open on the office computer or on your laptop, in Chrome, Edge or Firefox. It works on the voice notes arriving in the chats open there.
A day of viewings, in order:
- Install it and leave the tab open. The extension is on the Chrome Web Store. Installing takes less time than the first voice note in the queue.
- Switch on Automatic transcription. New voice notes become text without a click. By the time you are back from a viewing, the queue is readable.
- Set up Per-chat settings. The chat with the overseas owner can arrive translated, while the team group keeps transcription only. One rule per conversation.
- Ask for a Bullet-point summary on the long ones. The 4 min 12 s note becomes three lines: figure, condition, deadline.
- Fill in Glossary and names your way. Development, street, building manager, estate — the right spelling once, and done.
- Keep what matters in Notes per contact. Wants ground floor, fine without parking, 620 if the kitchen units stay: it sticks to the contact, visible only to you.
- Copy with 1 click and paste into the CRM. This is the step that erases the 40 seconds per note from the sum above.
- Find it later with Search inside voice messages. Ctrl+K, or Cmd+K on a Mac, type kitchen, and the passage comes up, grouped by conversation.
- Export the chat transcriptions as a .txt file when the deal becomes a folder and you want the negotiation history in one place.
- Schedule messages and files for the floor plan you promised to send at eight in the morning, while you still remember, at ten at night.
Where this stops
The part no advert mentions.
- It needs the computer on. The extension lives in the browser. A whole Saturday out on viewings, laptop shut, produces no transcription at all. The voice notes stay in WhatsApp and can be transcribed later, when you open the chat.
- There is a daily quota. The Free plan gives 10 transcriptions a day and audio of up to 15 minutes. Starter goes to 60 a day and 30 minutes. Pro goes to 150 a day and 150 minutes. In launch week, a large book of clients runs past Free.
- Proper nouns come out wrong. Developments, estates, surnames, street names: transcription stumbles. The glossary fixes what repeats, not what showed up once.
- The text is a note, not a document. It is there to jog your memory, log the CRM and brief the team. Offers, counter-offers and conditions are formalised in writing, in whatever instrument the deal requires. The 620 in this piece is illustrative, invented to explain the flow.
- It does not replace the call. A buyer who sent two minutes of complaint about the parking space wants to be called back, not answered in text. Reading first only means you call knowing what it is about.
Seven days to try it on your own book
One week of viewings is enough to measure this. Every ZapVox sign-up starts with 7 days of the full feature set, with no credit card. When the seven days end, the account drops back to the Free plan and nothing is charged.
Try it free for 7 days, no card: switch automatic transcription on Monday, work a normal week, and on Friday look at how many times you reopened a voice note just to check a figure.
In two lines
- The voice note belongs to the client, the record belongs to you. Between the two sits a stretch of the day nobody measures.
- On the phone, native transcription handles one note at a time. On the computer, where the CRM is open, an extension is what makes the voice note arrive as text.