Love Is Blind works. That is exactly the problem.

The pods are not the ridiculous part.

Love Is Blind: UK returned to Netflix on August 19, which means another set of strangers has just fallen in love through a wall. Watching any version of this show with the research in hand is a disorienting experience, because the element everyone calls absurd is the one with the strongest evidence behind it, and the element nobody questions is the one holding nothing up.

The pods are doing something real

Strip away the set design and a pod is a room where two people who cannot see each other talk for hours, take turns, and escalate steadily from small disclosures to large ones. That is not a gimmick. That is a procedure, and it has been tested.

In 1997 Arthur Aron and colleagues published the study now known as the 36 questions, in which pairs of strangers worked through three escalating sets of prompts, taking turns. Forty-five minutes later they were meaningfully closer than pairs who had made small talk. The findings that get less attention are the ones that matter most here: matching people on important attitudes did not help, expecting to like each other did not help, and being told to try to get close did not help. The mechanism was the structure of the disclosure itself. We went through what the paper actually asked people to do in the 36 questions.

Now scale that from forty-five minutes to ten days, remove every other channel, and give the participants nothing else to do. The pods are Aron's procedure with the volume turned all the way up. The intensity that viewers find implausible is the most predictable thing on the show.

The format's other premise holds up better than people give it credit for too. Sorting by appearance first is a genuinely poor filter, and when researchers fed more than a hundred self-reported traits and preferences into a machine learning model to predict who would want whom, the model could not do it. That is covered in machine learning cannot predict chemistry. A photo grid is not sitting on some superior science that the pods abandoned. Both are guessing. The pods at least guess after a conversation.

So: people can fall in love without seeing each other first, fast disclosure builds real connection, and appearance-led sorting deserves the skepticism the show gives it. On all three the format is right.

The failure is one step later

What the show then does with that feeling is where it comes apart.

Having reliably manufactured closeness, the format asks for a marriage decision, within weeks, and treats the strength of the feeling as evidence about the decision. Those are two different quantities, and the research is unusually clear that the second one cannot be read off the first.

In 2020 Samantha Joel and 85 collaborators pooled 43 longitudinal studies covering more than 11,000 couples and ran machine learning across the lot, asking what actually predicts whether a relationship is satisfying. The winners were not traits, demographics, values or preferences. They were qualities of the relationship itself: satisfaction, commitment, appreciation, and above all each person's perception that the other was committed. Those predicted around two to three times better than any fact about the individuals. We wrote that up in what makes a relationship last.

Read that list again with a pod in mind. Every item on it is a property of a shared history. Perceived partner commitment is not a trait somebody arrives with, it is a conclusion you reach after watching someone choose you repeatedly under conditions where they could have chosen otherwise. In a pod, nothing has cost anything yet.

So the variables carrying the predictive weight are not yet observable in there, and nobody has validated a way to read them that early. Say precisely what that is and is not: the 2020 study measured these variables in couples who already had a shared history, so it does not show they are absent from a pod conversation. It shows they were measured somewhere else, on people further along. The claim that a pod cannot surface them is our reasoning from the nature of the variables, not a result anyone has published.

The pods produce closeness, which is real and fast. The show reads it as compatibility, which is real and slow. Everything else follows from that one substitution.

That is the substitution at the heart of the format, and it is why the show keeps producing the same outcome. It takes the one thing about human connection that is genuinely quick and uses it to settle a question that depends entirely on the things that are not.

The funnel is the finding

You can watch the correction happen in the structure of each season.

Dozens of people enter the pods. A handful leave engaged. Fewer still reach a wedding, and some seasons produce none at all. Season nine of the US show ended with no weddings. Season ten sent seven engaged couples out of the pods and finished with two marriages. A minority of the couples who have married on the show are still together years later, which is a real outcome and a small fraction of the people who felt certain in a pod.

Every one of those narrowings is a couple learning something the pods could not have told them. Not because the pods were badly run, but because the information had not been generated yet. The engagement-to-altar attrition is not the show failing. It is the show working as an instrument, measuring the gap between how close two people feel and how well they will actually do, and finding it wide.

Worth adding the obvious limits. This is television, made by producers who need a season to be watchable, and the casting, editing and deadlines all serve that. It is not a controlled experiment, the participants are not a random sample of daters, and n is small. Nothing here is evidence about your relationship. The pattern is suggestive, not conclusive, and the show's real value is as an unusually vivid illustration of a mechanism the actual studies have already established.

What to take from it if you are dating

Expect closeness, not a verdict. A conversation that goes deep fast is producing exactly what the research says it should. It is a genuinely good sign and it is not a result. The feeling is real, and it is early.

Notice what a fast connection cannot contain. Whether someone shows up when it is inconvenient. What they do when you disappoint them. Whether their account of their last relationship matches how they behave in this one. None of that is available in week one, from any format, no matter how honest the conversation.

Do not read intensity as evidence. The strength of a feeling tracks how the disclosure was structured, not how well the two of you will do in three years. Those are separate measurements and the show conflates them for you every episode.

Give it the time the slow variables need. Not as a rule about waiting, but because commitment, appreciation and the sense that the other person is staying are things that have to accumulate before anyone can see them.

The part Bloom is built around

We think the pods identify the right mechanism and then hand it to the wrong moment, and most of what Bloom is follows from that reading.

Profiles are built to be read rather than glanced past, because disclosure only works when it lands on a person instead of a frame, and the free plan gives you a set number of likes a day, since a bounded day is the condition under which anyone has attention left for a real conversation.

But the piece that matters most here is what happens after. Prompts in Relationship Mode are Aron's mechanism pointed at the relationship you already have: reciprocal, escalating, taken in turns, on an ordinary Tuesday rather than under a deadline. That is the setting where the slow variables actually accumulate, and it is the half of the arc that almost every dating product hands off to nobody. Bloom is designed to be kept for that reason rather than deleted the moment it works.

Ten days and three years

The show's title makes a claim that is roughly true. Love, in the sense of a powerful bond forming between two people, does not require sight, and the pods demonstrate that every season.

The claim it never quite makes, and the one the format depends on, is that this bond tells you what the next thirty years will hold. On the evidence, it does not. It tells you that two people disclosed honestly to each other under conditions engineered to make that happen, which is a fine beginning and a poor prophecy.

Everything worth knowing comes after, which is the least televisable fact about relationships and the most useful one.

Bloom is pre-launch, starting in Utah along the Wasatch Front. The only thing to do today is join the waitlist and grow with us.