- A good reason for the revision. An agents' comments are the best; however, you could also say something about a writing professor/published author/critique partner having given you an idea/some criticism. Always frame in the positive, ie, I came up with this brilliant new idea! versus My manuscript had problems, but now it's better.
- Acknowledgment of agent's busyness.
- Polite and confident request to disregard the old version and check out the new one instead.
- Acknowledgment that the agent may have looked at the old version already.
- Expression of excitement at the possibility of working with the agent.
- A note that you've attached the revision.
- A thank-you and a promise not to do this again.
Is this a good idea? Strictly, no: I've had writers submit three, four, FIVE revisions before I'd even had a chance to comment on the first--and then, I'm always stuck with extra emails floating around. Sometimes it's not even clear which revision goes with which query which goes with which author--and I have to go on a hunt. It leaves me terrified that I could reject something twice, that interns will spend time reading something twice, that I'll never know which version is which. It's a mess.
Worse, I often can't see a significant change in the new drafts. A typo, for example, is not reason enough to resubmit: keep in mind that you're asking an agent to read your work again, and if we're going to do so, we want to feel like you've made changes large enough to change our decision.
That said, I think every writer gets one revision--IF they go about it correctly.
Here's a real revision note, accompanied by the revised manuscript, that not only left me better-than-annoyed--but pleased and intrigued.
If you're going to do this, always reply to the most recent email we've sent, something that has the full trail of our correspondence, so we'll know immediately who you are. If you can't find such an email, consider including your query, cut-and-pasted into your email, with the synopsis and manuscript attached.