The cornerstones of email metrics are open rate and click rate. As you might have guessed, the open rate is the percentage of recipients that opened an email and the click rate is the percentage of recipients that clicked on said email. So far, so good.
Now, for someone to click on your email, they first need to open it. Hmm, so we need a way to standardize what I like to call “click intent.” This is where the click-to-open rate comes in. This metric helps you understand what percentage of subscribers that opened an email also clicked on a link.
To calculate the click-to-open rate, divide the number of unique clicks in an email by the number of unique opens of the email. Express this number as a percentage. You should be able to get unique opens and unique clicks from your ESP dashboard. Depending on your ESP, the click-to-open rate may be something you’ll have to calculate on your own using the below formula.
click-to-open rate = unique clicks / unique opens × 100
A few additional metrics of key importance are your unsubscribe, bounce, and spam rates. The unsubscribe rate is the percentage of recipients that unsubscribed from your emails, and the spam rate is the percentage of recipients that marked your email as spam. Bounce rate, you guessed it, is the percentage of recipients who didn’t receive the email—the send bounced.
Unsubscribes are one of the most misunderstood metrics in marketing. Embrace them.
To put unsubscribes into perspective, recipients are kindly asking that you no longer email them. This will always be some percentage of the recipients, and oftentimes it will be a trivial sample—2% would be incredibly high. What an unsubscribe helps you avoid is a poisoned list: a list full of bounces, or worse—recipients marking your messages as spam. Bounces and spam rates contribute to your ability to reach the entirety of your list. There is no greater fear for an email marketer than getting blocked—even temporarily—by an ISP like Gmail, Outlook, Comcast, or Yahoo!. I’ll say it again: Embrace unsubscribes. The net gain is far greater than losing those addresses who do not want to hear from you.
One pro move is to allow subscribers to unsubscribe from specific campaigns (e.g., stop sending me your weekly newsletter—I still want to receive flash-sale alerts). Just know that the easier it is for someone to unsubscribe from your list, the less likely they’ll mark you as spam.
Pro tip: Allow subscribers to manage their communication preferences. Let them choose which emails they receive and how frequently.
Now that you know what to track, let’s talk about how we can improve each metric.
Improving open rate
The easiest way to improve your open rate is to A/B test your subject line. Another good subject line complement is the pre-header text. The pre-header text is the first [n] characters of the email—the value of [n] depends on each email client. The pre-header text will show either next to or below your subject line, again depending on the email client. Your pre-header text should complement your subject line to inspire the recipient to open the email and should be at least 40 characters long.
Email preview with subject line and pre-header text.
Another variable you can test is the from name—who the email is coming from. The sender of an email could be [company_name] (e.g., MailCharts), or [first_name] from [company_name] (e.g., Carl from MailCharts), or just about anything you’d like. Set this up in your ESP when creating your email.
The last variable you can use to improve your open rate is email send time—when is the email being sent. Your open rate will vary based on the day of week and time of day it was sent. You should test this and understand what happens when you send emails in the evening vs. morning vs. lunchtime. In a recent time-of-day test I did for a client, the difference between the lowest and the highest open rate was a 40% improvement. All emails were sent the same day and had the same exact content.
Also, if your ESP allows it, segment subscribers based on timezone—to standardize a “9 am” send time across the East and West coasts (and the rest of the world).
Quick aside: You can run time-of-day tests the same way you’d run an A/B test, except there’ll be as many variations as times you want to test. Make sure to schedule these tests at least one day in advance—allowing you to test early send times (e.g., 6 am) and not needing to manage the send process throughout the day.
Improving click rate
When it comes to click rate, everything in the content of your email is up for optimization: calls to action, imagery, copy, links, length of email … and the list goes on.
Here are a few general rules that are likely to boost your click rate:
- Make sure all images are clickable
- Use low-commitment CTAs. “Learn more” is often a better option than “Buy now”
- Make sure buttons and links stand out by using bright colors
- Give subscribers an incentive to click through
When analyzing your click rate, remember to normalize based on the open rate—use the click-to-open rate, as mentioned earlier.
A word on opens and clicks
At best, opens and clicks can inform the success of your email campaigns. And at worst, they can wholly misrepresent the performance of your program. Let me explain.
An email’s click rate and open rate don’t really matter if the email doesn’t achieve its goal—unless, of course, an open or a click is the goal. Most often, the goal is something else—it is a conversion. Conversions can take many forms—purchasing a product, completing a survey, or a user activity, such as completing a view of a video.
Here’s where opens and clicks may deceive us. While an email may have a 100% open rate, if nobody buys, then did the email succeed (assuming your goal is to sell)? My suggestion is to look at the numbers based on the entire funnel. If revenue is your goal, you should be able to calculate how much money your last email campaign drove. If user reactivation is your goal, you should know how many users successfully reactivated thanks to your email.
When analyzing the entire funnel, opens and clicks are just one part of it. Optimizing these top-of-funnel numbers can have a dramatic impact all the way down the funnel. But in isolation, it’s difficult to assess the efficacy of a program by looking simply at opens and clicks. (“Free iPad,” anyone?)
While email marketers must be wary of measuring performance based solely on opens and clicks, this is all too often the case with us in the email world—because obtaining conversion data is both difficult and time consuming. My advice is to prioritize getting conversion numbers, even if it requires signing up for a business intelligence tool such as Looker or Chartio. They’ll give you the leverage you need when requesting more resources, a raise, or anything else within the organization.
Improving conversion rates
While conversion rate optimization (CRO) deserves a whole ebook of its own, let’s cover some quick tricks you can use to boost your email conversion rates.
First, make sure you’re linking to the right place. When showcasing a product, your links should drive visitors to the product itself and not to the homepage. Always aim to link as far down the funnel as possible. In the example of a cart abandonment email, drive subscribers directly to their abandoned cart.
Second, automatically apply discounts. If you send an email promoting 20% off, don’t force subscribers to add the code manually. All this does is add friction, which reduces conversion rates.
Third, create some form of urgency. It could be that the promotion expires soon or that only a few items are left. Make subscribers feel that the best time to purchase is now; otherwise, it’ll be too late (without sounding spammy or sleazy!).
Calculating the impact of your campaigns
The best way to measure the impact of your email marketing campaigns is to have a holdout group (also called a control group). A holdout group is a set of users, usually between 2% and 5%, that don’t receive specific email campaigns. By creating a holdout group, you can compare how users that receive emails behave vs. those that don’t. Ideally, the group receiving emails does more of the thing you’re trying to achieve: stay past a free trial, purchase more often with a larger average basket size, refer additional friends to your service, etc.
When determining the impact of your campaigns, the attribution model you use plays a very important role—switching from a last-click attribution to email open paired with a 7-day window can paint a very different picture. Here’s the spoiler, though: There’s no right or wrong attribution model. You should look at your data through different models and understand what this means for your business.
Also, do your best to place dollar-based metrics around your email efforts. Paired with the holdout group, you’ll be able to point at the numbers and say, “Look, I generated an extra $1 million this year by sending these campaigns. Give me more resources. I need a graphic designer, a copywriter, and these other tools.” Otherwise, it’ll be difficult to get these additional resources to grow your email program.
Repercussions down the funnel
The other benefit of tracking dollar-based metrics is that when you run A/B tests, you’ll be able to measure the impact all the way down the funnel. Setting up and tracking A/B tests takes time and resources (design, copywriting, etc.). When a test wins, you need to know what the dollar impact is.
Other sources to help you improve your emails
The next time you’re looking to improve your email metrics, speak with your support team. Before launching a new email initiative, give them a heads-up. Once the campaign is live, ask them what feedback they’ve been hearing and look at recently opened support tickets. I’ve done this time and time again with my clients.
To give you some context, here’s something that happened recently. The MailCharts team launched a test for one of our clients where emails were personalized based on a subscriber’s industry. The goal was to increase the in-funnel sales conversion rate. Guess what happened?
The test failed. We looked at the numbers and couldn’t understand why. When speaking with the support team, they mentioned that some of the industries in which we were bucketing people were not fully accurate. This miscategorization was making some of the subscribers lose trust in our brand. Aha! Now that’s an explanation I can get behind. There’s no way we would have gotten this insight purely by looking at the numbers.
Another option is to talk directly with customers. Two popular options are to run a focus group in person or recruit participants for a phone call via email. I usually offer a $25 Amazon gift card in exchange for 10 minutes of someone’s time. Not everyone you reach out to will want to do this, but, as a rule of thumb, I expect a 30% answer and participation rate.
The last option you have is installing more tracking on your website. You can use a mouse tracking software such as MouseFlow or MouseStats. Additionally, if you use an analytics tool such as Amplitude or Mixpanel, you can add all the necessary events to understand exactly where people are dropping off in your funnel. Is it because subscribers are not opening your emails, clicking through to your landing page, not adding items to their cart, or not completing the checkout form?
Pro tip: Over 50% of emails are opened on mobile, so make sure your emails and landing pages are mobile optimized. Passing an identifier—like a unique token or a hashed email address—allows you to “identify” users after they’ve clicked through to your site and you’ve sent the appropriate identification events (no worries if this sounds foreign, you can ask your engineers to do this).
Implementing your optimizations
Before deriving any insight from your email campaigns, make sure you have statistical significance. Start at the top of the funnel, where you’ll have the most data (email open rate), and work your way down (checkout page conversions).
Every test you run should be based on a clear hypothesis. For each hypothesis, you should be able to run many tests to prove/disprove it.
Don’t simply change a button color because you want to make it “pop.” That’s not a hypothesis. Saying that the button does not stand out is much better. You can do many things to make a button stand out. Change its color, yes, but you can also add white space around it, make it bigger, etc. You should be able to come up with a series of tests for each hypothesis.
Handling low subscriber volume
Everything mentioned above works only if you have a large-enough email list. If your list has fewer than a few thousand subscribers, I suggest you focus on subscriber engagement vs. specific metrics. Ask questions to prompt a response. This will be much more insightful than looking at small numbers.
This is exactly what we do at MailCharts: As soon as someone joins our email list, we have an autoresponder that asks for their No. 1 email marketing challenge. We want to know why someone joined our list and what they’re struggling with.
These interactions help us understand who is interested in our product. Based on these answers, we’ve refined our marketing activities, landing pages, website copy, and even our product. You can do this only if you know your visitors: What do they need? What do they want? What are their fears? What do they want to learn about?
Email and content marketing: A match made in heaven
No matter which inbound activities you’re focused on (blog, podcast, ebooks, infographics, etc.), create a calendar outlining these efforts and plan your email strategy around it.
Inbound marketing plays really well with email—whenever someone reads your content (e.g., a blog post), your goal is to capture their email address. Once you have their email address, your goal is to get them to open your emails and return to your website to continue consuming your content. Over time, you’ll build trust with your subscribers, which then makes it easier to convert them into customers.
Also, your email list will often be the first wave of readers whenever you publish a new article or post. They’ll help you with the initial distribution needed to give any good piece of content momentum to succeed.
A note regarding older email subscribers
Generally speaking, older subscribers (e.g., those that joined your email list a year ago) are less likely to open your email than those that joined 2 days ago. It’s important to keep this in mind: If subscriber growth slows down, you may see “lower performance” in your emails (lower open, click, etc.). This doesn’t necessarily mean you’re sending worse emails today, it may simply be that your list is aging—meaning the average time since signup is increasing.
This insight heightens the importance of creating a welcome email or new-subscriber drip.
Quick aside: I have yet to come across an ESP that allows you to look at email performance based on subscriber cohort (ex: show me the open rate of subscribers that joined last month vs. those that joined 6 months ago). One day we’ll surely get this type of report.
Sample cohort report from Kissmetrics.com (an analytics tool). Imagine if you had access to this in your ESP!