How we screen astrologers

An astrologer who sells remedies will not be listed on bhagol. That was settled before there was an application form to turn anyone away with, and it rules out a large part of the trade, including careful people.

The rule is not a verdict on anyone's character. It is about who is paid for which sentence. If the same person finds the fault and supplies the cure, finding the fault is worth money to him, and nobody has to be dishonest for that to bend the work over ten years. So the category is decided in advance and printed where an applicant can read it before spending an hour on us. Turning somebody away for a reason published beforehand is a different act from turning him away because something about him did not sit right.

Two more categories are settled the same way, both checkable from outside.

Anyone who promises an outcome. A marriage inside the year, a case won, a visa cleared. Nobody has to guess at intent: the promise is usually printed on the practitioner's own page, next to a price.

Anyone whose first move on a stranger is fear. The message that names a serious problem and then declines to say what it is. The free chart that ends in a doṣa (दोष), a fault read in a birth chart, with a payment link underneath it. That script is most of the reason this company exists, and a practitioner who runs it is not somebody we can improve by listing him.

Character is not a thing I can screen for. Nobody can. What can be screened is what a person already sells, and what he does with an hour of work while we read it.

What gets checked

A screen here is a fixed set of checks, run against written rules, before anyone is listed. Six steps. Each one is a promise about what we will do, not a report on what we have already done, because none of them has run yet.

The price list first, with the public profile as it stands on the day of the application. It is the cheapest check and usually the most informative. A line for a stone. A rate charged by the minute. A guaranteed result. A free report that ends in a named fault. Any one of those closes the file. What he does elsewhere is his own affair. We read the list for the shape of the work, because that shape would arrive here with him.

Then a work sample, because claims about work are not work. Two readings: either past ones, with the customer's details removed and that customer's permission, or fresh ones written on birth data we supply. We read them against the rule our own transcripts follow, set out in what an honest reading looks like. Every line either names where it came from, the text and chapter, the school, the convention a number was computed under, or it carries the writer's initials, marking it as one person's judgement. A reading fails this test not by being wrong but by being unmarkable: if no line can be put in either column, the customer cannot tell what he is being asked to take on trust.

Third, the conventions. Which ayanāṁśa (अयनांश), the offset that fixes where the sidereal zodiac starts, which house system, which ephemeris. This is not a test of doctrine; the schools differ and are entitled to. It tests whether a practitioner can say which school he is working in, because a number handed over without its convention cannot be checked by anybody, including the person who paid for it.

Fourth, three questions a chart cannot answer: whether to go ahead with an operation, whether to fight a case, whether an illness at home will pass. We are looking for one sentence such as "this is outside what a chart can tell you", in whatever words the applicant prefers. Anyone who answers all three has told us what we needed to know.

Fifth, the agreement, signed before any listing exists. Fixed listed price and stated scope. No meter running. No remedy sold, in the session or after it. No forecast of events. No approach to a customer outside the platform to sell anything at all. The refund path honoured as written rather than as felt on the day. A signed rule turns a disappointment into a breach, and a breach has consequences attached.

Sixth, a real name, and a way to reach the person after the session ends.

We ask for no certificates, no lineage, and no count of years in practice. None of the three turns into the sentence somebody writes to a frightened stranger at nine at night, and that sentence is what I need to know.

There is no rejection rate on this page, because nobody has applied and nothing has opened. I could tell you what share I expect to turn away, but it would be a guess with a decimal point on it, which is worse than silence. When there are applications, the counts get published: how many applied, how many were listed, how many were turned away, and under which category. That is a later essay.

What the screen cannot catch

Four things, and I would rather set them down now.

An application is a performance. Anybody can write two careful readings knowing they are about to be read closely. The answer is not a better form. The reading does not stop at the gate: a share of delivered transcripts, with the customer's details removed, is read on our side for as long as somebody is listed, and every listed practitioner knows it. The knowing is where the value sits.

A transcript catches sentences and misses tone. The same sentence can be said kindly or used to press somebody, and where a session is spoken the customer is the only one who heard how it sounded. That puts the weight on whether complaining feels worth the trouble, which is the complaint path's problem, not the screen's.

What happens afterwards sits outside the screen entirely. Suppose the message arrives three days later, offering something at a friendly price, from a man who was impeccable for the whole hour before it. Nobody has done that to us yet; I am describing a shape, not a person I have met. No screen catches it in advance, so the rule carries the weight instead: any approach to sell you something off the platform ends the listing. I would rather hear about ten harmless messages than miss the one that was not.

And people drift. Passing a screen says something about the week somebody applied and nothing about the fifth year of a thin practice, when sharper phrasing starts to pay. So a listing carries a date and gets re-read on a schedule.

Screening will fail. The honest question is what happens on the day it does, and that machinery already exists, because it was built for a wrong date in an essay. The corrections page, at /corrections/, sets out what is owed when we get something wrong: a reply with the working, a fix that is never made quietly, and a dated note that stays. A reading that breaks one of the rules above is an error of that kind, not a customer-service matter.

A listing that ends because a rule was broken will say so. Not a name in a headline, and not a quiet disappearance either: the page carries the date and the rule that ended it. A platform that removes people silently is asking you to trust a screen you have never seen working.

None of this asks you to wait for us. Raman set out what a person can check in the room, in how to choose an astrologer. Every item on our list is a question you can put to an astrologer yourself.

The screen is written and nobody has walked into it yet. When the first application arrives we will read the price list before anything else, and if there is a stone on it, that is where the conversation ends.