Texting was supposed to replace talking. It replaced browsing instead, and the receipts show up in an unlikely place: what happens to your reply rate when you type “u” instead of “you”.
Texting was supposed to be the thing that finished off the phone call.
It wasn’t. Americans exchanged nearly 2.2 trillion SMS and MMS messages last year, the highest total of any year except 2020 — and over the same twelve months they spent more than 2.4 trillion minutes talking on the phone, according to CTIA’s annual survey of the US wireless industry. Messaging added roughly 42 billion messages on the year before. Voice added nearly 30 billion minutes.
Both lines went up. That is not a channel replacing another channel.
Something did get replaced, though. The part of buying that used to happen in the open — the aisle, the catalogue, the page you scrolled through before deciding anything — has been quietly relocated into a private conversation between one seller and one buyer. The window shopping is gone. The window is gone.
The revenue line nobody predicted
Meta now reports messaging as a business rather than a feature.
Paid business messaging on WhatsApp crossed a $2 billion annual run rate in the fourth quarter of 2025, and in the United States, revenue from ads that open a message thread rather than a web page grew more than 50% year over year, the company said in January.
Look at what that second format actually is. Someone pays for an advertisement whose only job is to start a chat. The click does not land on a product. It lands in an inbox, and a human being — or lately a model pretending to be one — takes it from there.
For roughly thirty years the entire discipline of online selling was about getting a stranger to a page and then getting them to a button. This inverts it. The page is skipped.
The conversation is the storefront, the pitch and the checkout at once, and it is unbrowsable by design: nobody can window-shop a thread they are not in.
Fraud found it before retail did
The clearest evidence that a channel converts is usually what the criminals are willing to spend on it.
In 2024, people reported losing $470 million to scams that started with a text message — more than five times the figure reported in 2020, according to the Federal Trade Commission, even though the number of reports went down over that period.
Hold those two halves next to each other. Fewer people filed complaints; the money lost multiplied by five. Whatever a fraudulent text was worth per attempt in 2020, it is worth several times that today.
The categories are almost boringly commercial. Fake package-delivery alerts. Bogus job offers. Fake bank fraud warnings, fake unpaid tolls, and the “wrong number” opener that begins as a message clearly meant for somebody else and ends, several friendly weeks later, in a fake investment account.
That last one is the tell. It is not a scam that abuses text messaging. It is a scam that is a relationship, conducted entirely by text, because that is now a format in which strangers will hand over money.
Every shortcut has a price attached
Which brings the story to the dialect.
Texting produced a compressed written language for practical reasons — 160 characters, a numeric keypad, a charge per message. Every one of those constraints is gone. The abbreviations stayed anyway, and they stopped being free.
Across eight preregistered experiments involving more than 5,300 participants, senders who used texting abbreviations were consistently judged less sincere, and they “received shorter and fewer responses and were less likely to receive contact information from the other texter,” according to research published in the Journal of Experimental Psychology: General in November 2024 by David Fang and Yiran Zhang of Stanford and Sam Maglio of the University of Toronto.
The effect held in lab studies, in a Discord field experiment, and in an archive of Tinder conversations spanning 37 countries. The more abbreviations, the less sincere the sender seemed.
The mechanism is effort. People read the length of a message as a measure of how much the sender cares, and they answer accordingly.
Which is a strange thing to discover about a channel that has just become a sales floor. In a shop, the transaction survives an indifferent clerk. In a thread, indifference is the product defect — and “u” is a two-character disclosure that the person on the other end is not trying very hard.
The problem with a storefront you can only enter one at a time
Every gain in a closed channel is paid for somewhere, and here the bill comes due before the first message.
A shop announces itself. It sits on a street, it has a sign, and people who were not looking for it walk past it anyway. A conversation announces nothing. It has to be opened deliberately, by someone who already knows who is on the other end.
Subscription platforms took this structure further than anyone. On OnlyFans, the monthly fee is frequently the cheap part and the paid conversation that follows is the actual business — the same shape as a click-to-message ad, with the advertisement removed and the relationship left in. The seller is not running a page so much as running an inbox.
That works right up to the point where somebody has to be found. A thread cannot be browsed, and platform search answers a name rather than a description, which pushes the finding step off the platform entirely: catalogues like SpicyCreator sort creators’ public pages by category so that a person can identify an account before there is any conversation to open.
Its usefulness stops at a fairly obvious edge: it holds what people have chosen to say publicly about their own accounts, confirms none of it, and cannot contain anyone who never announced themselves anywhere.
What the numbers do not cover?
CTIA’s survey counts messages carried by US wireless operators. Traffic inside WhatsApp, iMessage, Instagram and every other app-based inbox is not in that 2.2 trillion, which means the real volume of person-to-business messaging is larger than the figure and not measured in one place by anyone.
The FTC total counts only what people reported. Scam losses are under-reported for the obvious reason that reporting them requires admitting to them.
And the abbreviation research measured personal and dating messages, not commercial ones. It is reasonable to expect the effect to carry into a sales conversation. It has not been tested there.
None of it changes the direction of travel. The phone call was never the thing under threat. What text messaging replaced was the shop floor — and a shop floor you can only enter one person at a time is a different kind of business than the one it quietly retired.
