More than once I've read that you might want to gradually lower the volume at the end of a verse, so that raising it again at the chorus makes a bigger impact.
 
Any suggestions as to how many dB is a good reduction for those last 4 bars or so before the chorus?
 
Thanks in advance.

I don't typically raise everything. I raise surrounding areas.
for example: Vocals. I jack them up a notch 0.3db to help increase dynamics.
I find when you start cranking up the whole mix in spots, your limiting in the mastering stages becomes a more difficult job.
 
so instead, I play those guitar a little harder in chorus, bring up the vocals a bit.
this also largely depends on what kind of music you are doing.
if its a chorus that is slow and has a low dynamic range, this effect wont sound correct.
 
each engineer, producer uses a tool for a job.
similar to side changing. you wouldn't side chain a bass guitar to a piano. unless it was some odd effect.
just like using dynamics to raise things in mixes.
 
use it sparingly. raise your chorus up between 0.2 and 0.3db. that's not much but its enough to know their is something larger happening, what you don't want is someone noticing you turned up the volume knob.
you want it be heard and have someone simply say "that sounds good right there.
that's what subtle means.as soon as the Chorus ends, find a spot on the last measure to zip out.
begin the diminuendo 
(getting smaller, diminishing). Do Not drag it out. similar to a dress..lift it up, then put it back down.
find spots where the music breaks or hangs to lower / increase.
 
Work by these 2 methods:
Crescendo = gradually increasing in force, volume, or loudness
Diminuendo = gradually reducing in force or loudness
 
Experiment.

Personally I would not be lowering the volume anywhere especially towards the end of a verse because you might notice it and it may just feel like the brakes are being applied a little.
 
What I would do is mix slightly higher in rms value in the choruses. eg assuming I am working at say K-14 dB FS rms level for verses I would maybe mix just a little higher in verses eg -12 say for example.  The good thing about using a VU as well is if the verses are hitting 0 dB VU (which equals -14 dB FS) then pushing up 2dB will only show the needle just going up into +2 dB VU instead of hitting 0)
 
In mastering I would also be aware of what is going on there and maybe just push that limiter a little harder in the choruses. eg the clipping levels would not change anywhere in the song but I would be just nudging choruses up by about 2 dB rms to give them a little more power etc..
 
It also depends on how dramatic a chorus also tends to be.  Some chorues don't need anything but they still sound like a chorus because of what they are saying.  Other choruses can benefit from the rms level push though for sure.
 
Also as chuckebaby correctly points out sometimes I might just nudge the vocals and the drums a little in a chorus as well and leave the rest alone.
 

Just say NO to using a cheap trick like that to cover for a deficiency in the song.
 
The verse should be strong on it's own...and when the chorus comes in the chorus should have a higher energy level so that no tricks are needed to make the chorus stand out.
 
The chorus itself should make a bigger impact. If it doesn't, rewrite it until it does.  The chorus should deliver the goods, bring home the bacon, be the big payoff. Get the idea?
 
You might even want to incorporate a prechorus to start BUILDING as you come into the chorus.  Yes.... I will add more or perhaps use different instruments in the chorus that I use nowhere else in the song. Then I drop them out at the end of the chorus, but the volumes for the vocals and such remain the same. Let the song do the work for you.
 
http://soundclick.com/share.cfm?id=12024980  this is an example that hits all of the points I mentioned above.
 
There's no need to reduce volume at the end of the verse when the song does it's job properly.  If you wish to bump the levels half a dB or so IN THE CHORUS, that's up to you. Mostly I leave things at the same level and use the addition of instruments and BGV plus the "natural lifting action" of a well written chorus to make the chorus bigger than the verses.....
 
but that's just me.....

GH has it right.  The arrangement is what makes the Chorus pop, not the mix.  In fact, if the absence of a pre-chorus, I will often build at the end of the verse just a little.  Using or highlighting different instrumentation, a different rhythm pattern can be used to make the chorus stand out, but they are mostly things with the song writing and arranging.

Guitarhacker
Just say NO to using a cheap trick like that to cover for a deficiency in the song.

 
a Cheap trick ?
I learned that from reading Jack Douglas (The Producer of Aerosmith)
 
I understand there are different types of music and some may not benefit from different methods, 
But one can always become more informed by trying new ideas. 
Giving a small boost on certain parts can make things jump out. it can leave the listener saying "i don't know why...but this sounds great.
 
those are the magic tools. the ones when a listener hears it, he doesn't realize it, he just thinks...it sounds great.
 

C'mon guys, we all know dynamics is more than just volume, but volume tricks are a legitimate technique and this is a legitimate question. Of course, now that I've taken the moral high ground I'll give a vague answer just like everyone else. The answer is: do whatever it takes, as much as you can get away with without revealing the trickery.
 
The idea is to let the listener's ears relax, thus increasing its sensitivity to the dynamic boost that follows.
 
Sometimes it can be a dead stop, perhaps a quarter note but more likely shorter, perhaps 50 ms in duration. Other times a reduction of 1 to 3 decibels will do the job. Whatever you can get away with. 
 
More often it's a partial reduction, such as pulling out or thinning out one or more instruments. Some of my favorite tricks are:
- thinning the kick drum or ride hits, e.g. 8th notes to quarter notes for ride/hats
- dropping out rhythm guitars or automating a HPF to roll off the low end
- slight tempo reduction
- muting hand percussion, e.g. shakers
- simplifying the bass part, going from relatively busy to whole notes
- removing lead vocal double-track for the last line
- lowering the bass
- fast fadeout on the last note before the chorus
 
A great contemporary example is Steven Wilson's Hand.Cannot.Erase, which demonstrates a masterful use of dynamics even though the overall volume is fairly consistent. Listen to how he pulls out or thins instruments for dynamic effect. Just about every dynamic trick is employed, including brief silence, strategic lowering of volume and soloing an instrument right before a big crescendo.
 

 
 

My approach to gentle level rms shifting is not about making something weak better. It is actually adding dynamics to an otherwise great sounding track.  I have found that sometimes when more things suddenly come into a chorus if you keep the rms level of the chorus the same as the verse then the chorus can sound a touch quieter. 
 
It is a bit like the concept of when you are mastering an album of tracks that have big band tracks in it and quiet one or two instrument and vocal acoustic tracks.  If you keep the band tracks and the acoustic tracks the same rms level, the acoustic tracks sound louder.  So you have drop them by about 2 dB to make them sit nice with the band tracks. The band tracks then sound a little stronger and have more power as they should.
 
In this situation you can also lift sections of a song.  Sometimes it is nice to come the other way as well, when there is so much added a chorus might jump 4 dB rms all by itself so pulling something like that down 2 dB makes it sound better and it still has a little boost upwards at the choruses.  Only 2 dB now like it should be.
 
So slightly changing the rms level can be creative and really enhance the entry of sections that suddenly contain a lot of things going on.  It is actually adding dynamics in.  You can apply it in two places.  Mixing and mastering.  You can also leave it all alone in mixing and only apply some dynamic mastering in the mastering phase.  Doing it in both areas could be over doing it too.  You can also do it in the mix and just not tamper with the mastering as the verses transition into choruses and in most cases it will sound right too.  It can be as simple as automating the master fader.
 
When you work at K-20 in mixing you have got tons of room to also make rms level shifts like this without any side effects.  In live situations if the PA is cruising as it should be, a little level jump like this can and does sound great. 
 
I intend to control the SPL level in the audiences over the duration of the gig to quite a precise degree in my upcoming proposed live electronic music ensemble. It is easy to do and imagine the concept of sitting at one volume for most things but having other levels in reserve. Just pop to them at some dramatic times etc and jump to quieter levels for more intimate expression.  It will only make the gig sound better. The same concept can be applied to a recorded production.
 
It's time to make use of the enormous dynamic range we have available with such low noise floor now in digital.  It’s fun and can make the music sound way better.
 
Dave Steve Wilson sounds amazing. I wish everything sounded this good.  Another good trick too by the way not mentioned here I think is time signature changes. They can add great impact in a very subtle way.

these last 2 posts were outstanding. A bunch of great information and ideas.
another thing I forgot to mention was panning. it can really make things stand out.
it can also kill a good mix. but automating panning can sometimes be just as powerful in certain areas.
however if not done correctly, it can leave a sense of emptiness when things go from wide, to less wide.
 
for example, I can get away with having a dual lead panned  left and right, but when that lead is over, bring these instruments back to center can leave an empty sound in the sides.
I've found, overdubbing 2 extra guitars for solos parts works good. unless its a small part (a quick lick).
but this doesn't always work and can get old if its a whole album of over dub left/right pans.
needs to be done tasteful.
I guess that brings a lot of things in to perspective here.
as long as its done tasteful, anything can be done. that's where the magic happens. that's where experiments gone array become inventions of the future. im sure George Martin walked that path.
 

How much to lower volume just before chorus
 
 
OK, I'll do it for $120 (plus airfare)... tho honestly I don't see why you couldn't do it yourself.