A friend of mine recently asked for advice on different ways to approach the task of querying large number of agents. She asked things like: Should I query them all at once? Or should I query them in groups? Or should I query them serially, one at a time, and wait for responses? If I query them serially or in groups, should I go with my favorite agents first (and then second-tier agents next, then third-tier and so forth)? Or should I query third-tier first, second-tier next, and first-tier last?
Let me kill the suspense by skipping to the bottom line. I told her to query in batches, backwards (third-tier first, then second-tier, and finally first-tier).
Now let's look at the reasons why.
Querying 100 agents all-at-once is the worst strategy ever, IMHO. First, I believe that a pitch should always be considered a work-in-progress. You should always be open to the idea that your query can be improved. Say you write a query this month (without sending it to anyone), then go on vacation, then come back to the query. There's a substantial chance that when you look at the query through new eyes, you'll see ways it can be improved. Or you might see mistakes in the original query, as written. Either way, think how disastrous it would be to copy-paste the original query and send it to everyone under the sun. Any imperfections in it will be propagated to all agents, and then you've blown it. Arguably, at least.
Also consider the possibility that your original pitch is simply taking the wrong approach. That's something you can discover by sending it to 20 or 30 or 40 agents. If you really believe the query is the absolute best it can be and you've selected agents carefully (to match what they're looking for), you should get at least one positive response out of 40 agents queried. If you don't, chances are good your query is fundamentally flawed in some way. You should consider whether a total rewrite is called for.
I've queried magazine editors, book publishers, and others in the past, and I've found from experience that a pitch can nearly always be improved. A direct-marketing pitch (which is exactly what you're writing) is something you hone and sharpen incrementally and continuously, preferably on the basis of testing. Sometimes you decide that an entirely different approach would be better. Don't foreclose that possibility by spamming out your first-generation query to everybody at once.That would be unwise.
Sending out queries serially and waiting for a response from each agent before moving on to the next one is simply impractical. Let's say agents take two weeks to respond, on average. (Which is ridiculous, because the true answer is closer to four weeks.) If you're planning on writing to 40 agents, it'll take you 80 weeks to get to all of them. I don't know about you, but for me, that would be impractical.
The reason I used 40 agents in the above example is that in the real world, agents respond positively to only two or three percent of cold queries. If you think you're in that category, you need to reach out to 40 agents, because a one-in-forty success rate is a two-and-a-half percent success rate.
Sending out queries in groups is the way to go, IMHO. But even if you adopt that strategy, you should still not blindly use copy-paste, because (again) if there are imperfections in the pitch, you need to find them early on, not when you've already spammed everybody. That means you should read each query before sending it out. Believe me, after 20 or 30 or 40 re-readings of something, you'll find flaws. Unless of course you're undeniably the all-time best writer in the universe and can reliably turn out perfection on the first go.
Here's why you should send batches to third-tier agents first, then second-tier, then first-tier last. (Unless of course you have a recommendation from someone significant, like a bestselling author who already works with the agency in question. If you have that, contact that agency first.) Usually your first tier will contain a lot of top-flight agencies (in addition to containing the occasional boutique agency that just happens to be a special fit for your particular project). Top-flight agencies get phenomenal quantities of queries. They have more good material to choose from than bottom-tier agencies. Thus, your level of competition is very great when you go to a top agency.
The way to beat the competition (if you don't come with a recommendation that really counts) is to come at the first-tier agency with an offer already in hand. This usually gets the agency's attention.
So the strategy I would use is this: Send out your first batch of queries to bottom-tier agencies. If you get an offer of representation from one of those, tell the first-tier agency that you've already got an offer in hand but you want to consider the top-tier agency in question first, because you strongly prefer that agency and don't want to go with the other one unless you really have to. But don't reveal the name of the agency that you got an offer from, because the top-flight agency will likely assume that (since you're writing to top-flight agencies) you got an offer from another top-flight agency. And you do want them to assume this. You certainly don't want them to know that your offer came from some little-known one-person agency.
It's totally Kosher to pit one agency against another like this. I can tell you for a fact that this sort of thing is done all the time when agencies pitch books to publishers. They love to get an auction going. I did this myself once, many years ago. I had a firm offer (contract in hand) from Doubleday. Instead of signing the contract immediately (as most people would have done), I wrote to four other top-flight publishers, and in my pitch I told them I already had an offer from Doubleday. All four publishers sent me contracts immediately and begged me to sign. I had an auction going. (I finally went with McGraw-Hill.)
I hope this discussion has been useful for you. It was for my friend.