Online therapy really did solve a serious problem. Then the most private conversation in medicine met the advertising business.
Harder version: read this lesson at C1 level →
Look at each word. Say what you think it means. Then tap the card.
Tap a card to see the meaning.
Write one word from the cards in each gap.
1. I keep my diary — nobody else reads it.
2. The hospital has a six-month for that operation.
3. The doctor antibiotics for the infection.
4. Please don't this photo with anybody else.
5. He to call me and then he forgot.
6. Our sales for this month is very high.
1. private · 2. waiting list · 3. prescribed · 4. share · 5. promised · 6. target
Read each part. Then answer the questions before you read the next part.
Start with the part that the later scandals made easy to forget. The problem these companies wanted to solve was enormous, and it still is.
In most countries, mental healthcare is the queue that never gets shorter. Waiting lists of months for a first appointment. Country regions with no therapist within a hundred kilometres. Costs above what most households can pay. And a shame that stops millions of people from even joining the queue.
Estimates suggest hundreds of millions of people worldwide have depression or anxiety and receive no treatment.
So "make therapy as easy to reach as a food delivery" was not a cynical idea. For many people it genuinely worked: a counsellor within days instead of months, from your sofa instead of a waiting room, at a fraction of clinic prices.
Platforms signed up thousands of qualified therapists as flexible workers. A computer matched them to subscribers. And the whole thing — video, voice, and above all unlimited messaging — went inside an app.
Investors loved the shape of it: a huge unserved market and monthly subscriptions. The advertising went through the new personal channels — podcast hosts reading therapy adverts between episodes, influencers describing their own sessions.
Then March 2020 exploded the demand. Lockdowns created a mental-health emergency and closed the clinics at the same time. Regulators suspended old rules, including the rule that you must see a prescriber in person.
Money followed. One company reached the stock market in 2021 valued at about 1.4 billion dollars. A newer one passed a valuation of 4.8 billion within two years of being founded.
And now two completely different professional cultures were sharing one product.
Therapy's culture: fifty minutes, strict confidentiality, progress measured in months, and success possibly meaning the patient leaves.
The growth culture: sign-up funnels, conversion rates, engagement dashboards, keeping the customer.
For a while the tension was invisible. Growth was access, access was the mission, and the mission was going magnificently.
True or false? Click your answer.
1. The original problem these companies attacked was real and serious.
2. Demand fell sharply in 2020.
3. Therapy culture and growth culture measure success in the same way.
The collisions came quickly, each along a different line.
The platform had shared users' health information with advertising platforms, including Facebook and Snapchat. That information included email addresses, IP addresses, and answers from the first questionnaire — including the fact that the person was looking for therapy. It was used to target adverts and to find similar people.
It had promised confidentiality, repeatedly.
The fine was small — 7.8 million dollars, returned to users. The precedent was enormous. The intake form, the most sensitive document in the whole process, had been treated as marketing data.
The company admitted no wrongdoing and said the practice was normal for the industry. Uncomfortably, investigations suggested that was accurate: researchers found trackers all across mental-health apps.
The second was prescribing. One company had grown at startup speed with a simple flow: online questionnaire, video assessment, medicine by post — including, under the suspended rules, controlled medicines.
Reporting through 2022 described clinicians pushed towards thirty-minute assessments and high prescribing rates, and nurses describing targets that medicine should never have. Major pharmacy chains started refusing the platform's prescriptions. Federal prosecutors opened an investigation. The founder was removed, and the company largely stopped prescribing those medicines.
No court ever decided the model was criminal. The investigation and the pharmacies' refusal did the damage anyway.
Notice the mechanism. It is the same one as Volkswagen. No memo orders harm. The targets, the queue and the funding round order it indirectly.
The third was money. The company valued at 1.4 billion missed its numbers within months. Both founders left in late 2021. The share price fell by around ninety per cent.
True or false? Click your answer.
1. The company had promised that conversations were confidential.
2. A court decided that the prescribing model was criminal.
3. The problems were caused by one written instruction to do harm.
It would be wrong to end with only the scandals, because the honest record answers both ways.
Access really did improve. People in small towns got help. People who would never have walked into a clinic typed instead. Employers added services that did not exist before. For plenty of users, this was the thing that worked.
And the newest question is now arriving on top: AI. There have been cases of AI-written replies sent to users who did not know they were partly talking to a machine, and lawsuits claiming that companion chatbots harmed vulnerable teenagers.
The accountability problem is the same one the translation industry met — but here the stakes are at maximum. An unaccountable system is cheapest exactly where a mistake costs most.
So the living question is not "was online therapy good or bad?" The record measurably answers both.
The question is structural, and every health system, employer and investor is answering it right now, mostly by accident.
The most confidential conversation in medicine is now carried by companies whose economics reward scale, engagement and data.
So who sets the boundaries that the therapist's office used to set with its walls? The market that pays for access? The professions that carry the duty? Or the regulators, who arrive — as this case shows — about three years after the funnel?
Somewhere tonight, a person types their worst thing into a text box.
This whole case is about everything that happens next to those words.
True or false? Click your answer.
1. The lesson is simply that online therapy was a bad idea.
2. Regulators acted years after the problems began.
3. Mistakes matter more when the subject is someone's health.
This story is full of sentences that need extra information inside them: the company that promised confidentiality, people who live far from a clinic. That is a relative clause, and it lets you say more in one sentence instead of three.
who — for people. → "Therapists who worked flexibly."
which — for things. → "The questionnaire, which was very personal, was shared."
that — for people or things, in the common kind of clause. → "The company that made the app."
where — for places. → "Regions where there is no therapist."
whose — for possession. → "Companies whose profits depend on data."
The comma rule, which really matters: no commas = the clause identifies which one you mean ("the app that shared data" — that one, not the others). With commas = extra information you could remove ("BetterHelp, which is an American company, …"). Note: use which, never that, after a comma.
| Word | Use for | Example |
|---|---|---|
| who | people | People who cannot afford a clinic. |
| which | things (and after a comma) | The form, which was private, was shared. |
| that | people or things (no comma) | The company that broke its promise. |
| where | places | Areas where there are no therapists. |
| whose | possession | A company whose money comes from adverts. |
Write who, which, that, where or whose.
1. Millions of people need help never join the waiting list.
2. The intake form, contained very personal answers, was treated as marketing data.
3. There are rural regions there is no therapist for a hundred kilometres.
4. These are companies profits grow when users stay longer.
5. This is the same mechanism we saw at Volkswagen.
1. who · 2. which · 3. where · 4. whose · 5. that
Each sentence has one mistake. Say the correct sentence out loud, then check.
1. The therapists which worked for the platform were qualified.
2. The form, that was very personal, was shared with advertisers.
3. People who they cannot pay for a clinic use the app.
4. This is the town where I grew up in it.
5. A company who profits come from adverts has a problem.
1. The therapists who worked for the platform were qualified. (people → who)
2. The form, which was very personal, was shared. (never that after a comma)
3. People who cannot pay for a clinic use the app. (remove they — who is already the subject)
4. This is the town where I grew up. (remove in it — where already means in which)
5. A company whose profits come from adverts has a problem. (possession → whose)
Take your time. Short sentences are fine.
Tell me the story of online therapy. Six or seven sentences is enough.
What information about yourself would you never put into an app?
Where exactly is your line, and why is it there?
Faster and cheaper help reaches more people. Slower and stricter help is safer.
Which would you choose for your own country? Explain the cost of your choice.
The regulators arrived about three years late.
Is that normal? Can rules ever keep up with new businesses?