How to Write Progressive Metal - Part 1: Reverse Engineering

In part 1 of this series, you'll learn how to perfectly, profoundly understand the songwriting process for any genre, and begin composing in any style you want, as if you've had years of experience.

Ultimate Guitar
Hey, guys! Kevin Goetz here again with another free lesson. This is the beginning of a series that will focus on learning to write your own progressive metal music. The first step to doing that, is a trick called reverse engineering. Note that once you've learned this method, you can apply it to any genre, not just prog, and especially not just metal. Reverse engineering is something of a primitive way to refer to the process of learning how songs are put together, by taking a pre-existing song and dissecting it. It's a process of discovery no different in principle to medical science being advanced by autopsy, witnessing how the parts and pieces affect and assemble the whole. This is, in my opinion, the fastest and most effective way to learn how to write your own music. The sci-fi nerd in me likes to think of it as the ability to assimilate the compositional style of a song, adding it to my own repertoire, by breaking it down to its fundamental level and understanding it on that most basic level. As a self-taught guitarist and a self-taught songwriter, I ended up testing out a number of different methods when I was really trying to get good at songwriting; studying each and every aspect of music theory I could wrap my head around, loosely improvising into Audacity through a USB mic, having jam sessions with my band and trying to pick out parts that sounded good, but in the end, I made the most progress in the shortest amount of time when I finally just downloaded a tablature editor and decided to write out each instrument, piece by piece, riff by riff, song by song, and I learned how to do this by observing those who did it before me. I started by transcribing, as best I could, a song that I'd had stuck in my head for a while at the time. It was the theme song to a '90s cartoon called "The New Adventures of Speed Racer." There were no particular criteria I used to make this choice, but looking back, I firmly recommend starting with something fairly simple and easy to transcribe. This is hard enough to do without complexity slowing you down. The first instrument transcribed was the drums. This established a framework for identifying the tempo and rhythmic syncopation of the rest of the instruments. I used the same scale and melodic intent as the original song, as well, but from here, I was able to begin to experiment. The guitars in the original song weren't doing anything particularly interesting, so I simply messed around with converting what I believe were half-note power chords into sixteenth-note notation, adding rests and single-note flourishes. I had to go back and add in more hits on the kick drum to sync up with this new riff better, but it was a huge improvement. I let the bass guitar sync up with this riff, and then added some kind of background padding synth; I think it was probably a choir, if I remember correctly. I believe I just set the synth up playing long, sustained triads with the same root as the emphasized notes of the guitar riff. So for instance, if those emphasized power chords were F5, G5 and A5, the choir would play F, G, and Am. The vocal melody actually took me the most work; I had to do some Googling to find out what notes and octaves were most favored by singers for different scenarios and sounds, e.g. Male "shriek" notes typically fall in the fifth octave, while the fourth octave tends to be more solid. Once I had that sorted out, it was just a matter of playing around with the chosen scale, matching notes up with the rhythm of the guitar riffs and the power and volume required to sing each note, and then matching lyrics to the end result. Over time, by "re-writing" more and more songs and assimilating more and more melodies and construction types, I assembled a songwriting library of sorts. You don't have to rip off songs in order to do this; it simply helps to understand how the song's original style was composed. And once you know that, it's yours. Want to learn how to write a neoclassical shred piece? Transcribe a few and modify them to your tastes. Chances are, you'll come away from that with a healthy grasp of harmonic minor runs, sweep picking, and diminished substitutions. Transcribe and modify a few jazz pieces and you'll come away a master of the ii-V-I. Understand what gives a catchy hook its catchiness by transcribing and modifying that pop song you can't get out of your head. So, download your tab editor (I'm partial to Tuxguitar since I've never wanted to pay for Guitar Pro), pick a song, ideally an easy one, and start learning how to compose like the pros! Leave a note in the comments if there's anything you'd like me to clarify. Note that this series' companion video playlist is updated almost three times as frequently, so if you get impatient, head on over here:
Until next time, see ya!

27 comments sorted by best / new / date

    I get what you guys are saying about this "just being a story," but it's actually really helpful if you want to write a certain style and aren't sure how to get started. Some people may really be into ragtime blues and aren't quite sure how to start writing in that style. By breaking it down and figuring out what makes ragtime different from other blues styles, you can start to develop your own ways of making that happen (which is what makes one songwriter of a certain style different from another songwriter of the same style). Besides, if he told you every step to take to write your own music, wouldn't he be writing it for you?
    Exactly! This was never intended as a magic formula; as I'm sure you all know, there IS no magic formula. And as I said repeatedly, this is the very first step in an ongoing series. You have to see how the instruments are put together before you can...put them together. One last thing: The "step #1," etc., expectation is actually kind of...covered. I gave you guys the order in which the transcription makes the most sense in the way that it's written out.
    The method is ok, but this lesson wasn't about reverse engineering too much, but about yours story with a song. Try to analyze some good lessons also, and try to be more informative Anyway, this is the similar idea which I used for years, before I understood theory, I wrote songs based on what was before, but.. this is somewhat just a beggining.
    This didn't actually do anything but tell a story on how to transcribe a song. That's cool and all, but in a lesson, we expect tips and things like "Step # 1". This was just, "This is what I did. It should work for you".
    In a genre like progressive metal where the goal is to break the rules more than any other genre something like this is close to useless. But if you really wanted an extremely general tip it would be to start getting used to other rhythms except 4/4.
    Seeing that this is part I we can hopefully expect the following lesson to live up to the name (progressive metal). This article has little and less to do with a genre-specifif approach, thus I found the title misleading and the article a little disappointing. However, I do understand and to a certain degree agree with what you say about "reverse engineering" and using software, I have found it a great addition. A new and wide world of sounds and structures.
    Don't worry, it'll get there. This curriculum is quite large, with over two-dozen installments planned out. I want to be sure to cover every last detail in its own time, you know? You can check out the Youtube playlist linked above to get an idea of some of the topics coming up.
    I kinda do something really similar. I believe it is important to listen to lots of music, and isolate little passages that you like, and understand what it is that makes you like that passage... a certain inversion, a certain harmony, etc... All in all its not about being original, its about being authentic. Make you own mental library of resources, be handy with Guitar Pro, and spend lots of time with guitar in hand... You might be surprised. REMEMBER IT IS FUNDAMENTAL TO LISTEN TO MUSIC IN AL ANALYTICAL WAY! (if you are gonna use this method!)
    Kevin, I didn't look at your other article because to be honest, I'm not a progressive player. Love a lot of it, but it's not my thing as a player. Just wanted to post a comment specifically to you based on this single article. I think the biggest problem with this article is that most of the people reading it have probably never actually taught before. They're looking for instant gratification. "Here's step 1, 2, and 3. There! Now you're a great songwriter!" From a learning perspective, it can be frustrating to start with something that seems simple or "not what you're looking for." I just wanted to say from someone who has spent time teaching people how to play, improvise, etc. that it seemed like a good place to start to me (even if nobody else seems to think so.)For anybody reading, think about it. If you want to learn how to improvise guitar solos, you have to learn your scales first. At least if you want to understand what you're actually doing. Does learning your scales automatically make you a good improv shredder? No. But it's a hell of a start. You start with the basics of what you want to do. Then you build. Period. If you understand it, then you can create something original based on that knowledge. Otherwise, you're only recycling what you've learned to play. And maybe you don't need a computer TAB program to analyze and understand what's going on. But if you're smart enough to analyze every intricate detail of a complex song just by listening to it, be smart enough to respect that some people just need to write it down. Or better yet, go put your superior skills to good use and start having people pay you to teach them how to do it. I would bet your idea of what constitutes a "useful" lesson would change drastically. I know mine did. Now look what you guys made me do...I'm that guy that writes douchey rants that should have been way shorter about how my perspective is better than yours when I actually don't even know you. Damn these forums.
    Thanks so much; this is a very encouraging perspective I have to admit that the initial reaction to this was very unexpected. This curriculum has been well-received locally by both my students and other teachers, so I found it strange to see such a negative reception here. Your insight about lack of personal teaching experience as a culprit is dead-on, though, I suspect. It's the same with audio engineering: Bands come into the studio and expect things that are literally impossible, simply because they lack the experience with the craft to realize its impossibility. I suppose that's something of a trend.
    yes, break down your favourite music into its various parts and study exactly how it is composed together, thats fine. But I hope you don't teach people that writing music using tablature software is the way to go. It completely removes the human element that is so often missing from technical songs. It comes across to me that you are ok with that, "start learning to compose like the pros"... I think many pros would be horrified to see kids designing songs on a computer. IMO.
    I don't think you've actually talked to a pro then. I know for a fact that John Petrucci used software to compose the drums for A Dramatic Turn of Events.
    I'm afraid that's simply not the case, Decembero. The list of names of those who do at least some of their composing with software, solely in the progressive genre and off the top of my head, includes John Petrucci, Jeff Loomis, Misha Mansoor, Tosin Abasi, and Michael Romeo. Outside the prog genre, a well-known composer who's amazingly good at retaining the "human element" is Tuomas Holopainen. Tab editors provide the simultaneous stimulation of audio and visual centers as you utilize the decision-making part of the brain setting up instruments, engaging more of the brain and stimulating more of your thought centers. Nothing but benefit comes from that.
    That said, I agree that there's a certain degree of sterility in prog metal right now, and in this series, I will address it. In my experience, that stems more from the mixing stage; editing the drums to the grid down to a fraction of a second, tracking riffs at half speed, deleting random notes out of a string of sixteenth notes, filtering the bass guitar out WAY too high...spend some time over at Andy Sneap's forum to hear some people get onto some REAL rants on this subject.
    I was referring to your point on composing, not over-engineering after recording. But yes that's what I mean, prog metal bands are a dime a dozen, especially the growing 'djent' scene, and many lose that human touch, occasional imperfections etc. It's just less organic. And programmed drums just have less soul in my opinion, however I'll readily admit it's better to do that if you are on a budget and plan to record with a real drummer, saves a lot of money and studio time. Mansoor did that (or at least a hybrid of synth drum loops and manual chopping and changing) and Periphery are far from sterile. Maybe progressive is the genre, the music, AND the writing process.
    Definitely, I agree that the tendency to compose everything using the same tropes, so to speak, is a recipe for sterility, and as I said, I will address that over the course of this series. The innovation will be in the hands of the musicians, however. I'll supply the tools to go beyond the standard, but I can't stop people from employing those cookie-cutter tricks. It's easy to do, so, sadly, many will do it.
    I guess I'm just not seeing what you guys are complaining about. All the info is there, in a clear, ordered set of paragraphs. This has already helped several of my students before I put it online; I know for a fact that it works. I guess you just expect a more scientific writing style, and are thrown off by the more literary approach? I'll keep that in mind.
    I dont know what you are trying to build into, and maybe the next article might elaborate. But as of now this article is somehow useless. The method isnt wrong but the fact thats its suppose to help in composition (that too of a progressive song) is doubtful. Maybe if this article came later in the series after a few other well built explainations, would it be somewhat more enlightening. After all, prog is a genre where an individual sound is very important (See DT clones).
    It seems to work for my students. I started at this point, as do they, and it provides enough comprehension of the musical building blocks to begin understanding how to put them together. It is, in fact, common to intuitively comprehend the use of one of the "standard" techniques such as odd time signatures after applying reverse engineering to a song containing one. This is the starting point because it enables the quickest knowledge gain out of all possible starting points, giving you a base on which to build your repertoire of techniques by UNDERSTANDING them, rather than simply USING them.
    I can't really see what's progressive about reverse engineering an already existing song - I mean, the whole point of being progressive is to progress, is it not?
    I believe what they're wanting, being a music lesson, is something a bit more concise and a bit less subjective. On another note, isn't "progressive" music written by individuals who are well aware of musical conventions?
    Again, I'm not seeing the subjectivity. I suppose for the rest of the series I'll dispense of the first-person narrative, as it appears to throw off the step-by-step, objective formula that is indeed there. As to your question: Yes, generally it is. This is an entire series of lessons, by the end of which, yes, all the theory tricks will have been explained in very effective ways. This is a complete curriculum with which my students have seen very rapid growth.
    KevinGoetz, it's actually not a particularly good start to listen to the bands you mentioned. Why? Because there are dozens of bands ripping off those sounds for the sake of sounding that way. And everything you seem to think is needed (modulations, odd time signatures, etc.) will be covered in further lessons.
    Hey man, not sure if you're going to read this comment but I really liked this lesson. I don't know what all the bitching was with these but I liked it and I'm going to check out more! Thanks!