This post is being republished from the Real Insights Archive.

You never get a second chance at a first impression, as the saying goes. The subject line of your email is your message’s first impression. When it works, it’s the start of a beautiful relationship, or at least a click-through. When it doesn’t, it chips away a little bit at the trust and engagement you’ve built with your subscriber over time. So getting the subject lines right is mission-critical in the email business. Improve yours through testing. Here are 3 ways:

1. Preview Pane Test

You write your subject lines so that they are read in their entirety, so it is important to make sure that all of the words actually show up in your subscribers’ inboxes. To do this, send blank messages from your regular email account to several of your own accounts, using your message’s subject line. View the message in the preview panes of as many email clients as you have access to: Outlook, Mac Mail, web-based mail clients like GMail and the web interface for your business email, and mobile devices. (I’m no accountant, but this practice may give you the opportunity to write off that iPhone you’ve wanted for a while. Or better yet – expense it.) Your goal is to make sure that all of the subject line is visible, or that it’s not truncated to the point of incoherency. The Preview Pane test will quickly teach you the merits of brevity (if Twitter hasn’t already), and also to keep all the trigger words at the beginning of the subject line. Don’t overlook your sender name as part of the preview pane inventory as well. If your message comes from “Association Annual Meeting” you don’t need to include “Annual Meeting” or “Association” in your subject line.

2. A/B Test

You can also test multiple subject lines using Real Magnet’s A/B test feature. One way to do this is to split your list in half and try different subjects to each half, keeping the rest of the message exactly the same. Make sure that what you are testing can be isolated and replicated. For example, testing the order of words in your subject line might yield different results in your test cells, but it is not learning you can extend to your next message. In this case, you’re not really testing; you’re just hedging your bets. Instead, try testing concepts that you can repeat in different messages, like including company or personal names in the subject lines, including words like “new” or “today.” If your list is big enough, roll out A/B testing to small segments and then use the winning subject line on the large remaining portion of the list. How large does your list have to be? It depends on your open rate. Ideally you would like at least 100 opens for your results to be statistically relevant, and the testing segments should not take more than a quarter of your list. So if your open rate is 20%, each of your test cells should be 500, meaning your list should be at least 4000 names. Testing to larger cells provides statistically more relevant data, but reduces the number of subscribers left to benefit from the winning subject line. Try a few different cell sizes to see which yields the most confidence-inspiring and actionable data.

3. Integrity Test

It is true that the objective of your subject line is to entice your subscribers to open your email. But that is not it’s only objective, and it certainly is not the objective-at-all-costs. Like every other part of your email program, one objective of your subject line is to deepen your relationship with your subscribers. The pitfall that some subject lines fall into is to successfully compel a subscriber to open with a tantalizing headline that doesn’t truthfully represent what is actually inside. I call these Cosmo subject lines because the practice has been perfected by the quintessential glamor magazine. A article with the tempting title “Dare to go Bare at the Beach!” may very well be about pedicures. Magazine readers may grow fond of this parlor trick, but your subscribers will tire of it. Be sure that your subject lines accurately telegraph the message content, even if you think you could get a better open rate by being clever.