Boosting clicks: new results and insights
Latest posts | Feed | | By Mark Brownlow
In a previous post I noted how you might exploit the fact that not everybody opens every email. Adding links to articles or offers from the previous email might catch a few bonus clicks from those who missed the original.The concept worked well in one newsletter issue, but there were still three outstanding questions:
- is the CTR boost sustainable across multiple emails?
- do "old" links cannibalize clicks from the rest of the email?
- how can we get even more out of this tactic?
Each issue featured a "What you missed..." section down at the bottom of the current issue with short links to the last issue's landing pages. (See an example.)
Does the CTR boost hold up?
How many extra clicks did we get on each link by featuring it in the next issue's "Missed" section:
Jan 12th newsletter:
Article 1: +6%
Article 2: +33%
Jan 26th newsletter:
Article 1: +1%
Article 2: +8%
Article 3: +2%
Article 4: +23%
Feb 9th newsletter:
Article 1: +24%
Article 2: +18%
Article 3: +18%
Article 4: +117%
Looks like the click boost is holding up.
Are these links stealing clicks from newer links?
Only a proper A/B test would tell us about cannibalization of clicks from elsewhere in the email. But if these "missed" links are genuinely producing additional clicks, then we'd expect the average number of links each recipient clicks on to rise. Let's see...
Before implementing the "what you missed" section, an average 0.47 links would be clicked for each recorded open. The equivalent figure for the three issues featuring this section? 0.61.
Looks like we are getting additional clicks out of this.
How can we exploit this tactic better?
The best results came from Feb 9th. Interestingly, the Jan 26th issue had some deliverability problems.
It's obvious when you think about it. The more people likely to miss a particular email, the greater the value of featuring links from that email in your next send.
So you might use this tactic in the next email sent out after a deliverability problem (assuming you solved the problem). Or following any email that got an unusually low open rate. For example, it went out the same day the President got inaugurated - when people weren't paying attention to their inbox.
Is there a footer effect?
Note that the very bottom link always gets the best click boost in each of the three cases.
If people scroll down the email and don't find anything to click on, do they then "leave" your email by clicking on the last meaningful link you offer?
Something to think about. And this would tie in to the idea that your email needs a clear call-to-action at the top and at the bottom.
Is there a more links = more clicks effect?
Another issue to consider: are "links you might have missed" nothing more than a particular application of the idea that the more links you offer, the more clicks you get?
Here's a graph showing how the number of unique links clicked on per recorded open matches up with the total number of links in the emails (for the last ten newsletters):

Perhaps some correlation between the two, but not a huge one?
There's likely a tradeoff. More links means more choice means more chance people find something they like. But...more links means more clutter and content and the more chance people give up and move on to something else. There's a sweetspot somewhere I suspect, depending on your email model and audience.
Three final points:
1. This tactic works for me. It doesn't mean it will work for you...test it out first.
2. Analysis is good. Dig into those numbers: there's so much your campaign reports can tell you.
3. At the end of the proverbial day, all the above is just icing on the cake. You want more clicks? Target better.
Tags: metrics, email marketing, email newsletters
Permalink | February 16, 2009 | 2 comment(s)
Get posts like this: as an RSS feed | biweekly email | via Twitter
2 Comments:
Hi Mark,
We always use this technique in our internal news letter it works fine we find atleast 10-20% of subscribers clicking on "previous issue" link...
We are planning to test it with B2C mailling....
Chaitanya.
By chaitanya, on
17 February, 2009
Hi Chaitanya,
Excellent: thanks for sharing those numbers.
By , on
17 February, 2009
Comments closed during migration to a new blog platform in early May



