Putting together a good musical arrangement is an art unto itself. There’s no end to the number of options that present themselves to artists and producers who choose to wear the arranger’s hat in the studio; even those who’ve been at it for a while are always looking for new ideas. In that spirit, here are (in no particular order) 8 quick suggestions of things to think about and watch out for when putting together a musical arrangement.
A lot of material gets recorded during the initial tracking and overdubbing phases of production, but that doesn’t necessarily mean that it all has to be used in the final arrangement. Often the most musical results are achieved when the producer/artist picks and chooses just the parts that work together best for a particular song and arrangement, instead of trying to use everything that was recorded or programmed.
That may mean not utilizing some excellent performances or cool sounds, and that can become complicated when there are different people involved (disappointed band members or collaborators). But even if a particular performance is nicely crafted or especially well-played, if it doesn't fit in the project at hand then it’s better to cast it aside than try to force a square peg into a round hole (musically speaking). Good ideas will often return in another form at a later date anyway, and the current project will be better for not forcing things.
When artists and producers are working on a group of songs there can be a tendency to rely on the same instrumentation throughout the project. This may make sense if it contributes to maintaining a band’s consistent “sound”, but it still can be a good idea to vary the instrumentation up a little here and there, for some welcome variety. That doesn’t mean wholeheartedly replacing a group of guitars or synths with a string quartet—simply changing one or two instruments slightly can add a little extra something to a group of similar tracks.
Even subtle changes can be helpful—for example, swapping one of several acoustic guitars for a mandolin or the like; trying an upright bass instead of electric (when appropriate); using a Rhodes or Wurlitzer in place of an acoustic grand piano. The extra variety of tones over the course of several songs (or even in different sections of the same song) can draw the listener in anew, and help to keep things sounding fresh.
Even studio novices know to keep the lead vocal (or lead instrument if that’s the case) up front in the mix, but the same concern may apply to the arrangement as well. It’s important to insure that other musical parts don’t step on the lead, detracting from the lyrics or melody, especially if that melody is the main hook for the song. A part can sometimes get in the way even if it’s not loud enough to interfere by volume alone.
A background riff (counter-melody) may contain a note or two that clashes with the lead, either pitch-wise (like a stray passing tone that might sound fine on its own but is dissonant when heard under the main melody) or rhythmically (like a syncopation that just doesn’t work against the lead). Sometimes a producer may be focussed so much on the lead that these flaws can go unnoticed, so it’s important to occasionally take a step back and listen to the interactions of the parts with fresh ears.
Many “one-man band” producers and artists have to wear all the hats in the studio, but whenever possible it can be a good idea to delegate some aspects of the production to another, for an alternative or fresh perspective. A talented multi-instrumentalist may be perfectly capable of playing all the instruments and all the parts in an arrangement, but that doesn’t mean the song wouldn’t potentially benefit from bringing in another player here and there, who may bring a different take (no pun intended) to his part(s). We all tend to fall into musical habits (characteristic riffs, familiar rhythms) and even one or two tracks from guest artists can provide a bit of welcome stylistic variety.
By “doubling” here, I don’t mean the familiar mix techniques of playing the same part twice or duplicating a part and delaying the copy slightly, to make it sound like two players. I mean doubling in the more traditional musical (orchestral) sense—having two different instruments play the same part in unison. In orchestral arranging this has been long employed as a way to coax additional timbres from a fixed orchestral lineup—when two instruments play in unison, their harmonics and overtones combine and interfere, producing different tonalities that neither is capable of on their own (i.e. flute and oboe; trumpet and sax, guitar and piano; etc). Add to that the inevitable variations in timing and pitch and you can have a much more colorful tonal palette without having to deviate too much from a particular instrumental lineup.
Many arrangements start off light and then gradually build to a higher level of musical intensity—needless to say, in the appropriate musical genres this can be a very effective way of holding and building the listener’s interest throughout a song. But all too often, when this technique is used, the buildup comes too quickly, leaving nowhere to go (in terms of musical intensity) for the rest of the tune, which can end up leveling off and sounding anti-climactic as a result.
Vocalists sometimes do this (especially on those voice competition shows), turning to elaborate flourishes and dramatic high notes long before the melody has been established well enough for the listener to have a handle on it (which is essential to those flourishes eliciting the intended emotional response when they do come along). Arrangements can suffer from this as well. A track may start off lightly (say, guitar or keys and voice) and then the full band comes in, but if that full bands enters full blast on the first chorus, the next two or three choruses may fail to provide the same jolt, again making for an unfortunate anti-climax. If that’s happening, a better approach may be to have a slight energy bump on the first chorus (full band but not at full-tilt), and a slightly bigger second verse and chorus (extra rhythm part or two), saving the full-tilt entrance for the bridge or solo, of course remembering to leave at least a little something for the last chorus(es) to bring at least some extra energy to the end of the song.
Comping is a standard arranging technique in the modern studio: multiple takes are recorded, and the best bits of them are extracted and combined into a final composite track. This can be done with any parts, but it’s probably most commonly done with vocals. But, like many modern studio tools that can be employed in search of the best performance (in this case the idealized perfect vocal take), a little may be a good thing, but too much of a good thing can sometimes be bad for you—comping is one of those tools.
Some producers, in search of vocal perfection, end up cutting the various takes not only into lines and phrases but into even smaller bits, words, even individual syllables (!) This can indeed create a technically perfect vocal, but it may sacrifice some of the musicality in the process. Many vocal tracks have an overall arc to the performance, as the singer’s tone and intensity ebbs and flows in keeping with both the lyrical content and the musical dynamics of the song. But few vocalists do this exactly the same way every take—their performances typically evolve and vary somewhat over the course of the session.
When a comp is created from very small bits of several takes, any sense of arc—overall musical dynamic—present in individual takes can easily be lost, resulting in a technically perfect but musically sterile track. An alternative is to comp more sparingly—maybe find the best take of the bunch, and then use comping to correct and enhance the musical arc of that particular performance, swapping out bits and pieces in larger chunks (when that works musically) and only applying small corrections (syllables, letters) when really needed, careful to preserve the overall arc and unique character of the take chosen as the basis for the comp.
On the same note as the previous suggestion (re comping), it can sometimes be a good idea to avoid too much “perfection” in general. Now of course this suggestion may be somewhat genre-dependent; there are some musical styles that depend on rhythmic perfection (i.e. dance music), where full 100% quantization is one of the defining elements of the genre, and other tools like pitch correctors (Auto-Tune) may be deliberately applied specifically for the effect they are capable of creating while applying perfect tuning to vocals.
But for many other musical genres, especially those styles that would utilize (or at least emulate) live musical performances, it’s often a better idea to let a little imperfection in and preserve the performers’ individuality than to correct everything but lose that all-important feel. Approaches like partial quantization (less than 100% timing correction) and using pitch correction only on short phrases (rather than whole tracks) can go a long way toward preserving a little of that human touch.
And on that note, I’ll wrap this up. When putting together a musical arrangement it’s always a good idea to step back now and then and consider other options and alternative approaches—hopefully some of these suggestions will prove useful from time to time.