That said... as mcurry commented on this post -- I disagree with the need to show up at the agent's doorstep with an offer. It is quite likely having an offer in hand from a reputable publishing house will increase your chances. And there are certainly agents who will definitely take the so-called easy commission. I did it. Once. And then I vowed to never do it again. I made myself a solemn promise that I would always read the work prior to offering representation. In that particular case, my reasons were selfish. The author and I did not get on well, and I really was not a fan of their book, and it made the whole thing rather more work than it should have been. For both of us. I can't speak for authors everywhere, but I imagine they would prefer an agent who was also enthusiastic about what they were writing -- not just about how much money they could pull down. For my part, this certainly comes in handy when/if a client hits a plateau in their career, or even worse, a backslide. I can't speak for agents everywhere, but I like to have something else to sustain me when the commissions aren't pouring in.
Here's my advice for getting a good agent:
#1 Write a good book which also happens to be saleable. (Corollary: Not all good books are saleable. Unfortunate, but true. Further corollary: Write a bad book which happens to be saleable - not recommended, but it's been known to work. Or... sell your soul for literary fame and fortune if it's what will make you happy.)
#2 Do the research. Teresa mentions the AAR as one of her first links. It's a loosely aligned group of agents who commiserate (yes, I chose that word deliberately) on the state of the industry, and try to do good works for author-kind. And they also have the much-touted Canon of Ethics, which was established in an attempt to provide guidelines for respectable agent behavior (like not charging reading fees). Because anyone can become a literary agent. They can print up the cards and stationery and go to it. But there's no guarantee they'll have any idea how to go about it in a way that benefits their clients in the long run. Most agents I know join an already established agency in the assistant position and work their way up - sort of an apprenticeship tactic. That's what I did. (Or they're an editor who's hopped the fence.) So -- do the research. Go hunting on websites (like those mentioned in Teresa's post). Look in Jeff Herman's guide or the Writer's Digest guide. See if you can find out who represents your favorite authors or authors who write similar types of work to your own (many authors have links on their websites to their agent's website and/or list them in the acknowledgements pages of their books). Attend conferences and listen and ask questions. Make a short list of agent targets and then take due diligence in the crafting of your query and/or pitch.
#3 Be prepared. Getting published is hard. Sometimes even harder than writing. No, really. I have a great respect for those people who have the stick-to-it-ness to finish a manuscript. That's a huge investment, and many authors turn back before making it that far. But - beware ye who choose to continue the quest. Not to rain on the parade, but rejection is the norm. One of the publisher guides we're listed in asks about our rejection rate: most agents in that guide list a 99% as we do - few have anything better than a 90%. In many cases, I've heard the old axiom - you can't get a publisher to look at your book without an agent, and you can't get an agent without having been published (one of the other popular versions of this saying being you can't get a job without experience, you can't get experience without a job). It may often appear to be true. But it is not entirely true. There are few publishers these days who officially accept unagented work - this can severely limit your options as an unagented writer, but if you seem to be having no luck in the agent arena, may be a way to go. As for the agents, it *is* tough (but not impossible) to get one without having that deal-in-hand or having been previously published. It's also possible to be published and have an agent decline to represent you. I just looked at my client list and I can confirm that I took on at least one brand new writer last year - no short story sales, even. I also have turned down published authors this week.
#4 A bad agent is worse than no agent at all. Much repeated in many sources. If/when you get to a point where an agent has called you and offered representation, take a deep breath and make sure it's really what you want. Any agent will not do. Don't be afraid to ask questions. Most reputable agents will have no problem with answering ones like these. Most agents will not tell you particulars of deals they have made -- those details are proprietary information belonging to the clients they represent. I only add this one as a personal aside, because I once had an author demand (and I mean, demand!) to know this before I had even read their material. More vague items (like, which publishers an agent has done deals with) are fair game, imo.
#5 If at first you don't succeed, try again. I have clients on my list who have previously been rejected by me. In fact, the first sale I ever made was for one of those. And it's still happening today. (On the other hand, don't invest in so-called rubber manuscripts - the kind that just bounce back with tweaks within weeks. Reconsider steps 1-4 in each case.)
A further note on rejection - remember it's not about you, it's about your book. A lot of writers seem to take rejection very personally, and it's easy to do since they've poured so much of themselves into writing the book they're putting out there. But, the truth of it is, that, in almost every case, the agent doesn't know who you are. They're making a judgement based on what's on the paper. And just that. To the author, it's one submission out of a few, and to the agent, it's one out of the 100 that arrived that week. In any case, keep in mind there are many reasons your material may not fit what an agent is looking for, including the one where they are completely oblivious to your future potential. Here are some of the more common ones:
*Not written well enough. It may be okay or fine. But, these days, most agents want more than that. And further to this, there are a large number of submissions which are not spell-checked or proofread sufficiently. Or, disappointingly, where the writing doesn't even make the "okay" grade. This includes manuscripts that cannot follow basic (and readable) formatting. I'm not wed to any one font, or anything - I just want to read it and not have it hurt my eyes.
*Not the right kind of book for the agent. This goes back to the research. We get a fair number of queries for non-fiction material, but it says on our website that we are an agency for novelists, and that we specialize in fiction. This indicates to us that we were not chosen wisely for this submission, and that we're probably not even on a short list.
*The work isn't different enough or is *too* different. Agents need something to sell. Even if a writer has reached the stage where they have a pleasant enough mastery of language, the work might be too derivative. And editors are always talking about wanting something "fresh." Conversely, the work may be so far into left field that it would also be quite difficult to sell. Both of these obstacles can be overcome by discovering the mysterious quality of "voice."
*Just don't like it. Remember that agents are, at the heart and soul of it, readers -- who happen to be well-informed (one hopes) about what sells and doesn't sell. But - just as you might not be a fan of certain popular fantasy series, that agent may find others dull or uninteresting. It's hard to enthusiastically pitch something like that. In that case, the short list probably will yield someone who *is* into your kind of fiction. Otherwise, it was likely one of the other reasons. This last one is tough. And not fair. But you have to like it, or even better, love it, to be the right agent for it.
Caveat - just *one* agent's point of view, and we are all far too individualistic (I think we have to be) for everything to hold true in every case. But, taken generally, I'd like to think my statements above are fairly accurate.