In this lesson
Climb a density ladder from verse to chorus
Count the parts a listener can pick out in each section, give every section its own step, and let new parts enter on the downbeat.
Updated · 5 min read · Lesson 2 of 20 in Arrangement & Groove
In short
- Listeners hear the step between sections more than the size of any one section, so each section needs its own rung.
- Count parts by what a listener can pick out: a stack playing one line fuses into one part, and past about three similar parts people undercount.
- Bring new parts in on the first downbeat of a section, because entries are noticed more easily than slow fades.
In this lesson (8)
The static arrangement loop
You write a beat and build it up: drums, bass, chords, a lead. When you listen to the arrangement, the verse and the chorus feel the same size. You add another layer to the chorus and the mix gets muddy. You add a riser and the transition still feels flat. The song plays like one long loop with a vocal on top.
Nobody planned the density. Every part plays everywhere, so no section is bigger than any other.
Why it matters
The ear measures the step
A listener hears the difference between sections more clearly than the size of any one section. If every section has the same parts playing, there is no climb, and the chorus cannot feel like the top of anything. Managing density means deciding which parts stand down in the verse so the chorus can be the highest rung.
The count that matters is the number of parts a listener could pick out and follow, which can be far smaller than the track count. Four synths playing one line in unison count as one. A bass and a kick that always hit together may count as one. A pad holding long chords under everything counts as one, and often as less.
The science
Grouping and the limits of counting
Bregman (1990) showed that the ear groups sound into streams by how the parts behave. Sounds that start at the same moment and change together are heard as one source. That is why a stack of layers playing one line fuses into one thick part, and why your density count should follow lines and roles, not tracks.
There is also a limit to how many parts a listener can keep apart. Huron (1989) played musicians polyphonic textures of similar timbres and asked them to count the voices. They were accurate up to three. When a fourth voice joined, accuracy fell sharply, and the usual mistake was counting too few. Different timbres make separation easier, but the lesson for arrangement holds: past a few parts, an extra part adds mass and more than it adds a line anyone follows.
The same study found that voices entering were noticed more easily than voices leaving. That is the case for a ladder with clear steps. A part that enters on the of the chorus is an event the listener notices. A part that creeps in across the last bars of the verse is barely noticed, and neither is the section change it was meant to mark.
Try it
The density count test
0 of 6 steps done
Each boundary should now feel like a step up, and the chorus downbeat should land as the biggest step.
Common mistake
Giving every section the same weight
Letting every part play all the time feels like it keeps the energy high, but a section that never changes stops sounding big and starts sounding constant. It also leaves no room to climb.
The second mistake is a slow slide in density. If parts fade in gradually throughout the verse, the section boundaries lose their impact and the chorus arrives on a ramp instead of a step.
Takeaway
Assign each section a density step
Give every section its own rung on the ladder, and count rungs in parts a listener can hear, not tracks. Keep the verse lean, bring new parts in on downbeats, and let the chorus be the one place where everything plays.
Masking
One sound making another harder to hear, usually because they share frequencies at the same moment. A loud sound also masks what follows it for a short time (forward masking). The quieter sound is still there; the ear stops separating it.
3 questions
Check yourself
Sources
Show all 2Hide sources
- Bregman, A. S. (1990). Auditory Scene Analysis: The Perceptual Organization of Sound. MIT Press.
- Huron, D. (1989). Voice denumerability in polyphonic music of homogeneous timbres. Music Perception, 6(4), 361-382.
Next lesson · 3 of 20
An intro should tease the hook, not play it allAn intro that plays the full hook with the full beat leaves the verse as a step down and the chorus as a repeat. Tease the hook through a filter instead.Previous: Draw the energy curve before you mix

Written by Virzy Guns
Songwriter and producer, founder of Virzy Guns Production. Now building HealingWave.