Learn more from your click reports: CTA and article distribution

The last post demonstrated how past campaign reports might hide a whole host of useful information on optimizing your emails.

We used my own informational newsletter (featuring article headlines and teaser summaries) to explore how link number and spacing might impact click rates.

Let's use those same campaign reports to look at call to action and the order of content.

Call to action

There's a sort of seductive simplicity to finding the right call-to-action. For example, compare the click-to-open rates for different types of calls-to-action across all teaser texts:

• "Find" CTAs: 7.84% e.g. "find out here"
• "Learn" CTAs: 6.59% e.g. "learn more"
• "Read" CTAs: 5.00% e.g. "read on" or "continue reading"

Oh joy! If we just use the words "find out more" rather than "read more" in our links, we can expect clicks to jump over 56%. Hurrah!

If only.

The evidence certainly suggests a more positive and active CTA ("find" or "learn") works better than a negative, passive one ("read"). This is backed up by a test done by MarketingSherpa, who found "click to continue" pulled more clicks than "continue to article" or "read more."

However, there's a posse of ifs and buts riding close behind. For example:

1. The CTA needs to make contextual sense. It's no good using "find out more" as your link if the text before it doesn't indicate there's something worth finding out: the CTA works in tandem with the quality and content of the surrounding copy and images.

2. The color/design of your link also has an impact. Instead of text, you might consider using a HTML-based button. Back in 2007, Chad White reported on retailers having great success with that approach.

3. The winning CTA may change through time. AWeber also tested "images" versus "text" for the "continue reading" CTA and found the former performed much better than the latter...temporarily. After some weeks, the relative pulling power of each reversed.

4. Different CTAs resonate with different subscribers. When we talk about segmentation, we nearly always think of customizing offers or content to discrete recipient groups.

But why not customize the CTA to what you know of the recipient?

Might "impulsive buyers" respond best to "get it while you can" CTAs while "information seekers" need "learn more about this product" links?

5. The importance of the CTA may diminish as recipients become trained to recognize where the link is and what it looks like.

Long-term recipients may be less responsive to CTA changes than more recent subscribers.

Indeed, might it be worth mixing things up a little to keep attention focused on your CTAs. See, for example, the innovative CTAs used by Backcountry.

The message to take away is not that one particular CTA wording is always better than another. Your list will have its own preferences, depending on what exactly you want them to do).

The message is simply that words matter. Even something as apparently banal as the wording of a link taking people to a full article can make a big difference to responses.

[Related post: A better call to action in your emails]

Article order

Here's a graph depicting how click-to-open rate changes for an article, depending on which order the associated teaser text appears in the email (1st, 2nd, 3rd, 4th or 5th):



Articles featured first in an email get around three times the clicks of those featured in 4th or 5th place. No surprise there, of course.

Not all of that click lift is down to position alone: the most interesting articles always go first in the newsletter, so are likely to collect more clicks anyway.

Can we learn anything new from this trend?

Note the uptick in CTR for the last article.

This likely reflects how some people read through an email to the bottom and then move on by clicking the last link they see. I've seen this in emails before when looking at how people click on lists of links.

So consider reserving last place for the "less important but not least important" articles.

You can also think of the graph above as the standard response curve for a sequence of articles (in my newsletter).

This gives me a baseline on which to judge the relative "value" of new articles.

Say the first three articles in the next newsletter all get a 6% click-to-open rate. Looking at that stat alone, you'd conclude there was equal interest in all three articles.

But compare the pattern with our standard curve and it's clear that interest in article 3 is actually highest: it's level on CTR with the top 2 articles despite the click penalty it gets from being lower down the email.

Again, we're just scratching the surface here of what you might glean from your reports. Next up: the impact of teaser text paragraph size and numbers of lines...

More on metrics and copywriting | Tags: email marketing, cta, email design

 feed | email | Follow me on Twitter

The main website for everything email marketing | Introduction
Email advertising basics | Service listings | Sources of statistics
Advice on email design, delivery issues and list building

22 sources of inspiration
Outlook 2007 and HTML design
42 HTML email design resources

follow me on Twitter