C Major or A Minor? How to Tell Which Key a Song Is Really In

C major and A minor contain exactly the same seven notes. So do G major and E minor, and every other relative pair. That is why two key-finder apps can give you two different answers about the same song, why the chord chart you found online disagrees with your ear, and why you can lose twenty minutes arguing about a song you and your bandmate both already know how to play.

Counting the notes cannot settle it, because the notes are identical. Something else has to do the deciding, and once you know what that something is, the question stops being hard.

Why the notes cannot tell you

Every major key has a relative minor sitting three semitones below it. C major and A minor. G major and E minor. D major and B minor. F major and D minor.

The pairs are not similar. They are built from the identical set of pitches, in a different order. If you write out the notes of C major and the notes of A minor, you get the same seven letters both times.

So any method that works by tallying which notes appear, or how much energy sits on each pitch, is looking at information that genuinely does not contain the answer. A lot of automatic key detection works exactly that way, which is why it flips between relatives on songs that sound unambiguous to a human.

What separates them is not the ingredients. It is which note the music treats as home.

Three tests you can do by ear

The stopping test. Play through the song and stop dead on the last chord of the chorus. Does it feel finished, or does it feel like it wants to keep going? The chord that sounds like an ending is almost always the tonic, and the tonic names the key. Songs end on home far more often than not.

The hum test. Play the progression on a loop, then hum whatever note feels most restful over the top. Not the prettiest note, the one that feels like sitting down. Find that note on your instrument. That is your tonic.

The one-chord test. Play just the chord you suspect is home, over and over, then play the song. If the song sounds like it departs from that chord and returns to it, you have the right one. If it sounds like the chord is a waypoint rather than a destination, try its relative.

These three agree with each other most of the time. When they disagree, the song is genuinely ambiguous, which is covered further down.

What the chords tell you that the notes do not

The reliable signal is which chords carry the most weight, and where the music comes to rest.

In C major you expect C, F and G to do the heavy lifting, and G to pull back toward C. In A minor you expect Am and Dm to dominate, often with E or E7 pulling back toward Am. That E chord is the giveaway: it contains a G sharp, which is not in C major at all. A raised seventh like that is the clearest evidence you are in the minor rather than its relative.

Two other things matter more than they look:

How long a chord is held. A tonic chord that sits for four bars says more about the key than a passing chord that lasts half a beat. Duration is weight.

Where the progression lands. The last chord of a section is doing disproportionate work. A verse that keeps returning to Am is telling you something a note tally never will.

How Larka decides

Larka used to have this problem too, and fixing it meant changing what the decision is made from.

The old approach averaged the energy on each pitch across the whole recording and matched that against key profiles. That is the method described above that cannot work, because relative keys produce near-identical profiles. On a real guitar recording it would land on the relative minor of the right answer about as often as the right answer itself.

The current approach decides the key from the chords it heard, not the notes. Each detected chord votes for the keys it belongs to, weighted by how long it was actually held. A key gets a further boost if its tonic chord appears at all, and more again if the progression ends on it.

That is the same reasoning as the stopping test, done arithmetically. It is why the key you get now tends to match the one you would have picked yourself.

Finding the key of something you are playing

Open Listen in Larka and play. The chord you are on appears live, with the key and scale underneath, and the screen stops listening on its own once the key has held steady for a while, so you are not left holding a phone.

For something you want to keep, record it instead. The chord timeline shows the full progression with how long each chord lasted, which is the thing that settles relative keys, and you can scroll back through it rather than trusting a single moment.

If you want to see why an answer makes sense, the Scale & Key Info and Circle of Fifths screens show which chords belong to a key. Relative pairs sit next to each other there, which makes it obvious at a glance why they are so easily confused.

When the answer is genuinely both

Some songs really are ambiguous, and it is worth saying so rather than pretending every case resolves cleanly.

A progression like Am, F, C, G loops without ever committing. Depending on where you start counting and which chord the melody leans on, it reads as C major or A minor equally well. Plenty of pop songs live in exactly this space on purpose, because the ambiguity is what gives them their bittersweet quality.

When that happens, the useful question is not which answer is correct. It is which one helps you right now. If you are writing a solo, pick the one whose scale shape you know better, because they share every note anyway. If you are naming the key for a bandmate, pick the one that matches where the song resolves at the end. If you are transposing, either works, as long as you are consistent.

The theory only matters to the point where it makes the next decision easier. Past that, trust what you hear.