Related to: simile, snobbery, adequate axioms
Writing poetry is harder than it sounds, but easy to practice. Once mastered, the emotional impact it can add to even casual conversation makes it more than worthwhile.
There are two sides to writing an effective poem: the top-down logic of metaphor and imagery, and the bottom-up mechanics of rhyme and meter. Manage both, or they won't meet in the middle.
Let's say you're trying to lift someone as high up into the air as possible. You could kneel and cup your hands, but that depends on them playing along and stepping in the right spot. You could sneak up behind and kick them squarely between the legs, but that won't get them very far, or for very long, and they won't put up with such treatment more than once. Or you could build a framework, hang a swing, and give them a series of properly-timed pushes in the right direction.
A pure technical explanation (the cupped hands) depends on the willingness of the reader to slog through the whole thing, do some independent research to fill any newly-discovered gaps in their knowledge base, and generally cooperate. Without that minimal enthusiasm, the most brilliant insights can and will be dismissed as "too long, didn't read."
Aggressive proselytizing, at the other extreme, sacrifices content to put as few demands on the reader as possible. It is, accordingly, viewed as even worse than useless. An active offense, spam, something to be isolated and destroyed.
Taking the time to lay out a pattern, a rhythm, means that people will have some reason to keep reading even if they don't know exactly what you mean. It's a comfortable set of boxes in which half-eaten ideas can be stored for later, or a resonant frequency to carry information until the full message can be compiled.
Resonant frequencies can't create something from nothing. The Tacoma Narrows bridge wobbled for hours before finally collapsing; cumulative energy transfer from the wind over the course of those hours was orders of magnitude more than would have been necessary for, say, a controlled demolition with shaped-charge explosives. The advantage is that slow, steady sources are easier to find and easier to regulate. An appeal to people's tendency toward pattern-completion can be spread out over pages, instead of requiring a single perfect paragraph, and will not be consciously resisted by anyone who does not realize they are being persuaded.
So, setting up the rhythm.
Look away from what your words actually mean. Consider what they sound like, which syllables are emphasized, the flavor of your favorite phonemes. Then look back, and shuffle things around until they match.
Reinforce parallel points with parallel structure.
Tempting though it may be to advertise your sophisticated vocabulary, or even invent or co-opt exotic terms and phrases for the precise elucidation of some nuanced concept, simple english works good because we all polish it.
Long lines of big words slow down the flow.
There's more to effective writing than I could cover in one essay, of course. People spend years studying this kind of thing, and the few who really master it are paid accordingly. If it was easy, everyone would be doing it.
Metaphor and imagery are harder to explain.
Most of the time, your objective in communicating is to reduce ambiguity. You lay out your thoughts in order, package them securely, and hope that they survive however many translations it takes until they can be reassembled in the same order inside someone else's mind. A word that means more than one thing is, in that context, a navigational hazard; it has too many degrees of freedom, so more information must be included to constrain it, lock in the single intended meaning. Unintended potential interpretations are dangerous noise.
In compiling myth, you must cultivate multiple consistent interpretations. Ambiguity is, to a certain extent, your friend; whenever a word could mean more than one thing, that's a chance to save precious syllables, each meaning simultaneously developing a different level of interpretation. The catch is that, rather than using words as scalpels to meticulously dissect the issue one step at a time, you are juggling jagged axes and chainsaws. Every edge, every possible definition, must be sufficiently familiar to you that no disastrously unintended layer will emerge.
Different layers of the same message have different, but related, meanings. Each layer will be picked up by a different audience, or a different aspect of the reader's mind, and should be tailored for effectiveness accordingly. In my wisdom:foolishness :: tree:stones comparison, the feeling of validity comes from an appeal to the reader's intuitive understanding of botany and other basic physical sciences, typically developed since childhood.
In a sense, poetry is a perversion of public-key encryption. You take your message, in it's most concentrated form, and connect it to some nugget of knowledge or archetype. Without that cultural context, there's no signal, just patterned noise. The recipient applies their own private version of that archetype or meme, and treasures whatever insights can then be unpacked, thinking that it was a secret message intended just for them.
Wait, is getting really rich the standard? Because not all languages are equal in their ability to ring this bell at any given time. Right now, English is king -- its the lingua franca of half the world or more, and has a disproportionate number of populous rich countries using it as their first language -- America is conspicuously important in this regard.
Writers in other languages are at a severe disadvantage in terms of ability to cash in on their local popularity and influence. I understand that Russia, for instance, takes poetry very, very seriously; but its legal and economic system is not capable of giving its poets their honors in the form of money. For example, Yevgeny Yevtushenko is a big deal, and he's not starving as far as I can tell. But neither does he make anything like a tenth of what cheesy American hack (not hacker -- the bad, journalistic kind of hack) Dan Brown makes.
Speaking of Dan Brown -- he's a huge financial success, but he's the non-thinking man's version of Umberto Eco, who, from a financial point of view, made the extremely unwise decisions to be Italian and not to pander to the lowest common denominator among his potential readers.