Measuring Success: An Email Marketer's Guide to Common Metrics

Every year, email becomes a more important part of the marketing mix for companies large and small. That's great news for folks like you and me who think about the "killer app" more than the average person. But as the medium we know and love continues to gain traction, email marketers face more pressure to gauge, quantify, and optimize their efforts. The numbers are readily available, but what do they really indicate, what are industry averages, and how do your campaigns measure up?

Open Rates
Clickthrough Rates
List Attrition
Conversion Rates

Open Rates

While we believe the primary goal of any email campaign should be conversion (and conversion can mean lots of different things), open rates have long been considered the holy grail of email marketing metrics. However, according to MarketingSherpa's Email Marketing Benchmarks Guide 2009, analysts and experts are moving away from using open rates as a measure of success, using them instead simply as a benchmark to gauge the success of subject lines. Just last year, there's been a drop for the first time in the number of marketers who regularly track email open rates, indicating they're looking for better alternative measures for success.

What Is It?

Simply put, open rates are supposed to tell you whether or not an email gets opened by its intended recipient. By itself, open rate data can't tell you anything about how much or how little attention the recipient gave your email, so in many respects it correlates to the "impression" used in various forms of advertising.

How are open rates measured?

There are no set standards for calculating open rates; therefore, it is important to know how your ESP tracks results. The two primary methods for calculating open rates: dividing unique opens by either 1) the total number of emails sent, or, 2) the total number of emails delivered. The BlueHornet system uses the latter calculation, based on delivered emails, and provides actual numbers that indicate both unique and non-unique opens and the average number of times subscribers opened the email (total opens divided by unique opens).

How "delivered" emails are calculated: Like open rates, there are no standards on how delivered emails are calculated. For instance, the BlueHornet system calculates "delivered" emails as an actual number determined by subtracting all failures (various bounces and permanently invalid subscribers) from the total number of emails sent. According to the Email Experience Council, 11 percent of ESPs count every single message that was sent as being delivered. Only a small percentage of ESPs, including BlueHornet, actually have programs that track the number of messages that are filtered by ISPs or delivered to recipients' junk mail folders. Therefore, we see the importance of knowing your ESP's metrics, and tracking data the same way across all campaigns.
Understanding the difference between unique and non-unique (gross) opens:If you successfully deliver an email to 100 subscribers and 50 individuals open it, your system will show 50 total unique opens. If each of those individuals go back and open your email two more times, your system will show 150 total non-unique opens along with an average number of times the email was opened. Multiple opens are a very good thing. They indicate that subscribers are engaging with your message, and engagement is a critical part of every successful email marketing campaign.

What are industry averages for open rates?

Marketing Sherpa's Email Marketing Benchmarks Guide 2009 indicates an overall decline in B2C open rates in 2008. Most respondents surveyed saw open rates for B2C emails sent to house lists in the 20-25% range. B2B emails typically pull higher open rates than B2C emails. Sherpa reports an average 25% for B2B emails sent to house lists. Our own B2B and B2C benchmarks come in a bit higher than these averages, likely because of our clients' wide adoption of double opt-in (DOI) processes. Keep reading for more information on how DOI can help improve opens.

What else should I know about open rates?

There are a few important factors that can skew open rates low and high, which is why some pundits have predicted the eventual death of this metric:

Image blocking may be the most significant factor contributing to lower than anticipated open rates. If you've ever wondered why ESPs only provide open rate data for HTML emails, it's because images must display for an open to be counted as such. As you know, many email clients default to automatically block images from loading in HTML messages. If a subscriber has image blocking turned on in their email client, they may still see your email. But unless an image in that email is loaded, it will not appear in your open rate stats. That "impression" cannot be measured. Ditto for emails viewed as text in Blackberrys, cell phones, and the like.
Preview panes can cause your open rates to skew high. If a subscriber has their email client set up to automatically load images, your email may display in their preview pane even though they never technically open or look at your message.

How can I optimize my open rates?

Despite the many issues that can affect open rates, we encourage you to establish a baseline for your email campaigns and then optimize to drive ongoing campaign improvements. Based on our experience with clients across all verticals, these are the top three things you can do to optimize your open rates:

Switch to double opt-in: If you're not using double opt-in (DOI), make the switch today. Our promise? You'll likely see some drop in list size, but your open, click, and conversion rates will definitely increase.
Segment your database: Segmentation increases relevance, and relevance improves key metrics across the board. Easy entry points into segmentation include segmenting by stated preference and past purchase data. A favorite segmentation technique used by BlueHornet's top online retail clients is our segment by clickthrough feature.
Conduct A/B subject line tests: Subject lines can greatly affect open rates. Because conducting A/B splits on subject lines is so fast and easy, you should really be working this process into every email you send.

Clickthrough Rates (CTR)

Driving traffic to your website is often a primary objective for email marketers, and effective use of links in emails will do just that. But even if site traffic is not critical to your campaigns, it's still important to measure and optimize link clickthroughs because CTR provides valuable insight into the level of recipient engagement with your email. That's something open rates alone can't tell you.

What is it?

Alone, clickthrough rate statistics provide data on the links in your email that were clicked. When used in conjunction with web analytics software, you can continue to monitor engagement within your website.

How are clickthrough rates measured?

Again, there are no standards on how clickthrough rates are calculated, so it is important that you know how your ESP uses this data. According to the MarketingSherpa Email Marketing Benchmark Guide 2009, 43% of ESPs calculate the CTR by the percentage of emails opened that get clicked, 39% by the percentage of emails delivered that gets clicked, and 14% by the percentage of total emails sent. BlueHornet calculates CTR for each tracked link in your message as a percentage that is determined by dividing HTML unique clicks by unique opens. We also provide data on total clicks (HTML uniques plus plain text uniques) and the average number of clicks per subscriber (total clicks divided by unique clicks).

What are industry averages for clickthroughs?

According to MarketingSherpa, most 2008 clickthrough rates averaged in the 2.1-5% range for B2C, whether those clicks linked to newsletters articles, sales offers, or free offers. B2B CTRs averaged from 2.1% up to 10%.

What else should I know about CTR?

Ideally, when observing clickthroughs, you should compare total CTR, unique CTR and specifically what content is being clicked — this way you can begin to make improvements where needed and capitalize on existing successes. Also, don't forget to use clickthrough data for segmentation. BlueHornet users have the ability to automatically fill a segment with everyone who clicked a particular link, and market to them specifically based on that behavior.

How can I optimize my CTR?

Short emails that link to full content online often pull higher clickthrough rates than email newsletters that provide most of the content within the body of the email itself.
Try underlining the links in your emails. –This simple technique has improved CTR for many of our clients.
Test the location of links. In the email world, any content placed above "the fold" is considered to be in a prime position. But an old direct mail copywriting trick puts a call to action all the way down in the P.S.: section of a print sales letter, which is supposed to be the first thing a reader looks at. So don't make assumptions about what section of your email will pull the best CTR for your subscribers. Instead, test many different locations so you'll know for sure.

List Attrition

A common question asked by many is, "I am losing subscribers with each send. Is this normal?"

What is it?

List attrition isn't a metric in and of itself. It's a general term that refers to a reduction in the size of your database. Marketers are most often concerned with two common contributors to list attrition: opt outs/unsubscribes and invalid/undelivered emails. You can measure list attrition by monitoring opt outs, bounces, invalids, and delivery stats.

How is list attrition measured in my BlueHornet system?

In your system, you can measure list attrition by monitoring opt outs, bounces, invalids, and delivery stats.

What are industry averages?

Undelivered: Because the factors that contribute to undelivered emails span such a wide variety of causes, from subject line verbiage to brand reputation to the practices of emailers who may share your IP address, there is little value in reporting aggregate deliverability averages. Industry analysts do not typically provide data across ESPs either, since there is a lack of standardization in terms of how delivery rates are calculated.
Unsubscribes: Based on our own data, we feel an unsubscribe rate of higher than 2% is something you should be concerned about. We also recommend further investigation if you receive more than one spam complaint per 5,000 emails sent.

What else should I know about list attrition?

Don't automatically look at unsubscribes as a bad thing. A smoothly operating opt-out process lets subscribers manage their interactions with your brand while maintaining a favorable impression of it. And if someone's not interested in your email offerings, keeping them in your database won't drive conversions. It'll only bring your opens, CTR and other metrics down.

What you should definitely be concerned about are spikes in unsubscribes and undelivered emails. A spike in unsubscribes can indicate that the integrity of your list has been compromised, usually by importing a third-party list. But unsubscribe spikes can also let you know that you should adjust your sending frequency or increase the relevance of your messaging.

Spikes in undelivered emails can signal a blacklist or other problem at the ISP level. Learn how our Deliverability Management Services team can help by emailing SureSend@BlueHornet.com. Our team works on clients' behalf to identify and resolve the issue quickly.

Conversion Rates

Be wary of any hard numbers or percentages given for conversion. I may consider conversion a purchase, you may consider it a clickthrough to a custom landing page, and someone else may consider a conversion to be a subscriber picked up through a referral campaign. Nonetheless, you should define what a conversion is for each email you send. Then measure, track, and establish your own baseline metrics.

Taking Metrics for What They’re Worth.

In addition to the intricacies behind various common metrics, different methods of measurement are used across various ESPs, which makes your job—and ours—tougher than it needs to be. That's why we're committed to complete transparency when it comes to our metrics, why we continue to enhance the reporting data in our system, and why we call for and support efforts to standardize metrics across ESPs.

So know your metrics, watch them, and by all means, work to improve them. But be careful not to get caught up in comparisons that may not be "apples to apples." Look at the big picture and use industry benchmarks as a guide until you establish benchmarks of your own.