Kundli Milan (Match Making)

What Kundli Milan is

Kundli Milan — guna milan, ashtakoota milan, kundali matching — is the compatibility check performed before a Vedic marriage. Two birth charts go in; a score out of 36 comes out, along with a list of doshas.

It is probably the single most consulted calculation in Indian astrology, and it is worth being clear at the start about what it was built for. Guna Milan comes from a world of arranged introductions between families who knew nothing about each other beyond a name and a birth record. It is a filter applied in the absence of information — a way to ask “is there any reason to stop here?” when there was no other way to ask.

That is not the situation most people reading this page are in. The technique still tells you what the tradition says, and this app computes it faithfully. But a number derived from two Moon positions was never designed to adjudicate a relationship you already have.

The eight kootas, at a glance

The 36 points are eight kootas (categories), each worth a different maximum. The weighting is the tradition’s statement about what matters:

The total is a plain sum — no normalisation, no re-weighting. Each koota’s detail, its data tables and the exact rule the engine applies are in Ashtakoota.

Notice the shape of that table. Nadi and Bhakoot are 15 of the 36 points between them, and both are all-or-nothing in this engine: you score the full 7 or 8, or you score zero. Two charts can be strong in six kootas and still land near 20 because those two failed. That is the tradition’s weighting working as designed, and it is why the raw number swings so hard.

Everything comes from the Moon

Six of the eight kootas are read from the Moon’s nakshatra or its sign. The remaining two, Graha Maitri and Bhakoot, come from the Moon signs’ lords. The engine’s input for the whole 36-point score is four numbers per chart: Moon sign, degree in sign, nakshatra, and pada.

Your Lagna does not enter it. Nor does Venus, nor the seventh house, nor a single planet other than the Moon. This has a consequence people notice immediately: two candidates born the same morning score identically against the same partner, because the Moon barely moved between them.

This is not a defect in the software. It is what Ashtakoota is. It is also exactly why the app does not stop there.

What AstroShruti computes besides the score

The Match Making section is public — no Pro plan, no sign-in — and it returns considerably more than 36 points:

• Ashtakoota — the eight kootas, each with its score, both partners’ values, and the reason it scored what it did.
• Mangal Dosha — Mars checked from Lagna, Moon and Venus in both charts, with a couple-level verdict.
• D1 seventh-house promise — each chart’s marriage indication scored out of 100 from the seventh lord’s dignity, Venus and Jupiter, and benefic or malefic influence on the seventh house.
• Navamsa (D9) — the same seventh-house analysis on the divisional chart that classical technique treats as the marriage chart, plus a Vargottama note where a Moon holds its sign.
• KP marriage promise — the seventh cusp’s sub lord in each chart, read for whether it signifies 2, 7 and 11. See the KP system.
• Dasha timing — whether each partner’s running Mahadasha and Antardasha currently support marriage, and the next favourable windows.
• Porutham — the South Indian ten-check system, computed alongside Ashtakoota rather than instead of it, including the critical Rajju and Vedha checks that the northern method has no equivalent of.
• Remedies and a full Markdown report of every section above.

The advanced score out of 100 blends these: 35% Ashtakoota, 15% Mangal, 15% D1, 15% D9, 10% KP, 10% dasha. Its bands run Challenging under 45, Average to 60, Suitable with minor cautions to 75, and Strongly suitable above that.

The point of the blend is to stop the Moon from carrying the entire verdict. A pair with a mediocre Guna Milan but two strong Navamsas and a clean seventh house should not read the same as a pair that is weak everywhere — and under a raw /36 it would.

The score bands

The engine reports one of four verdicts against the 36-point total:

18 is the classical floor — the point below which tradition says the match should not proceed on gunas alone. It is worth knowing where that number comes from: it is half of 36, and it is a convention. It is not an empirical threshold, and nobody has demonstrated that it predicts anything.

The summary line the app writes follows the same logic: below 18 it says the traditional rules do not recommend the match; if Nadi, Bhakoot or a Mangal mismatch is present it says suitable with caution, review the doshas; at 32 and above it says excellent; otherwise, suitable.

We print these because they are the tradition and because you came here for them. We would rather you read the eight rows than the one number.

How to actually read a report

Some practical advice, offered as advice rather than astrology:

• Read the kootas, not the total. A 24 built on a Nadi zero is a completely different chart from a 24 built on scattered small losses. The total hides which one you have.
• Check whether a dosha was flagged as softened. The engine surfaces classical cancellation grounds where they apply — but it does not add the points back. See below.
• Look at the D9 and the seventh house. They read the whole chart, which the gunas do not.
• Nothing here knows either of you. Not your history, not your temperament, not whether you can talk to each other at eleven at night. Astrology’s own literature is more modest about Guna Milan than the internet is.

What we do not compute

Being straight about the limits, because this is a page where the limits matter:

• Cancellations are surfaced, never applied. This is the single most important thing to understand about the score. When Bhakoot Dosha is present, the engine checks the classical grounds — the two Moon-sign lords being the same planet, or being mutual friends — and if one holds it prints “Same Moon-sign lord” or “Moon-sign lords are mutual friends” beside the koota. When Nadi Dosha is present, it likewise notes same Nadi but different Moon signs, or same nakshatra with different padas. In both cases the koota still scores zero. The grounds are shown to you as information; the engine does not silently restore 7 or 8 points on your behalf. Many other calculators do, which is why our total can read lower than theirs on the same pair — the difference is a policy choice about cancellation, not a disagreement about the chart.
• No other cancellation rules. The classical literature carries many more cancellation grounds for both doshas than the two-per-dosha the engine checks. We do not implement them. If your astrologer applies one, the engine will not have.
• Nothing outside the Moon enters the 36. Ashtakoota does not look at Lagna, the seventh house, Venus, dignity, aspects, or divisional charts. The other report sections do; the score does not.
• Yoni’s middle values are conservative. Same yoni scores 4, an enemy pair scores 0, and everything in between is flat 2 — the finer classical gradations are not implemented, pending golden tests.
• No age, gender-role or life-stage judgement. The engine takes two charts labelled boy and girl because the classical rules are directional — Varna, the Vashya matrix, the Gana matrix and Stree Deergha all read differently in each direction. That is the tradition’s asymmetry, faithfully reproduced. It is not an endorsement of it.
• No prediction of outcome. Guna Milan does not forecast divorce, happiness, children, or anything else. It compares two Moons against a table.

Notes on this calculation

• Public — the whole report, no Pro plan, no login.
• The 36 points are a plain sum of eight kootas from each chart’s Moon.
• Verdict bands: <18 Not Recommended, 18–24 Average, 24–32 Very Good, ≥32 Excellent.
• Bhakoot and Nadi cancellations are reported, not scored. A cancelled dosha still costs its full points.
• The advanced /100 blend is 35% Ashtakoota, 15% Mangal, 15% D1, 15% D9, 10% KP, 10% dasha.
• Both charts are computed on your ayanamsha setting (Lahiri by default), and the report states which it used.