Posts Tagged ‘Email testing’
Posted by Bill Leming on July 15th, 2009
Several months ago I wrote a post about Re-Touch Efforts and Seeing Positive Results. In it I noted some unexpected success with re-mailing the exact same offer to individuals who neither opened nor clicked the original email they received three to four days earlier. What was surprising to me at the time was the fact that the re-touch effort resulted in the same response rate as the original. And it was surprising because the rule of thumb in direct mail was always that you’d get about half the response rate that you received in the initial mailing everything being equal.
But when you think about it, when are all things ever equal? They never are. Today is not like yesterday; next Wednesday won’t be like this past Wednesday, July 15 this year won’t be like July 15 next year. And that may well be the reason for email re-touches being much more effective than direct mail re-touches. Despite the differences, yesterday is generally more like tomorrow than a day 28 months from now just as an individual’s recent past behavior is more like is/her near term future behavior than their behavior at some distant point in the future.
I don’t have any empirical evidence but I think the reason email re-touches are more successful than direct mail re-touches is that they generally occur so much faster. You don’t need to wait 30 days for results or even two weeks for projected-results to determine which segments are responding to which offers using which creative approaches. Nor do you have to worry about production schedules and print queues. With email your questions are answered in hours. You not only know which segment(s) responded at what rate to which offer using which creative, you know that you can deploy the re-touch email to the right target in minutes with the winning offer tucked neatly into the winning design.
Since April we’ve had the opportunity to use these findings in combination with one another with some even more impressive results. Not only have we managed to equal the response rate’s of the initial offering, in more than one instance we’ve managed to triple it by combining the most responsive list segments with the most appealing creative.
In each of these scenarios we’ve chosen to hold the offer constant for purposes of simplicity and I think we’ve made a mistake in doing so. Inasmuch as the “offer” generally carries more weight in the email success equation than does “creative”, the next effort will be to measure the effect of quickly re-touching winning segments with winning offers using winning creative to see if we can’t raise the response rates even further. The trick will be to do that without getting so narrow and so microscopic that we find the one guy in 100,000 who’s ready to buy and no one else. Will keep you posted on our progress.
Tweet This Post
Posted by Bill Leming on April 13th, 2009
Several weeks ago I wrote about specific retouch strategies and their potential effectiveness in driving revenues. Since that time we’ve had a real-world opportunity to test retouch efforts to recipients who neither opened nor clicked on the initial promotional offer. While the objective of the initial effort was to capture more information about potential buyers and generate actual, measurable sales, the focus of the message was to enter a sweepstakes for a large screen, high-definition LCDTV and test our ability to further generate form completions with a retouch effort four days prior to the sweepstakes conclusion.
What happened was that the “last chance” retouch produced render rates nearly equal to the initial offering and click through rates only slightly less than the initial effort. More important from the client’s perspective was the fact that form completions (which was a major goal of the campaign) were actually higher than the initial effort and unsubscribe rates were lower. Certainly the reasons “why” this occurred has much to do with the high level of engagement the recipients feel towards the client.
Another perhaps equally plausible reason may lie in the fact that recipients of high-value sweepstakes with a limited time entry deadline may actually welcome the “last chance” message and see it more as a friendly reminder, which they appreciate and embrace.
Tweet This Post
Posted by George Palatine on April 1st, 2009
It is pretty typical for companies to use a functioning but not monitored email address as their “from name” in an email marketing strategy. Examples of this are the common “info@” or “sales@ “ structure. If a company has a large list or sends frequent mailings it is possible that this inbox could receive thousands of reply-to messages. Some of the more common include out of office replies and the dreaded “hard bounce” undeliverable, but what about the others -customers who do not pay attention to the “do not reply to this message” text, change of address notifications, challenge/response triggers, etc. Developing a plan to monitor these messages could provide some significant benefits.
A study of a B to B publisher done by Bamboo Cricket (a 3rd party service who can monitor/respond to your inbound email) found that as much as 6-8% of the sender’s outbound email volume resulted in an inbound response. As one would expect, most of these messages were non-essential, but over 9% were, in fact, high value communications requiring a human response.
If these results are typical, lets say you sent a message to 100,000 recipients, you would receive a “reply” from 6,000-8,000 of them, of which 540-720 would be high value communication. Would you let your call center or customer service department ignore 700 inquires? In this economy it is important to take advantage of every opportunity to build loyalty and retain customers.
There is no question that personally reviewing thousands of inbound responses can be a lot like panning for gold. You could spend all day with nothing to show for it, or perhaps you may find a few nuggets. Whether you develop an entire plan around this process or just do an Nth message sampling, you may find that “there is gold in them there replies. ”
Tweet This Post
Posted by Bill Leming on March 11th, 2009
As times become more difficult for many businesses, we see more and more pressure on Marketing to maximize both revenues and efficiency. The days of Senior Management adhering to a “belief” that a particular marketing tactic was efficiently generating sales without “proof of life” are long gone. And while Email Marketing typically falls into the high efficiency/high revenue producing 2 x 2 quadrant, the fact that it has largely been relatively inexpensive has in part led to email marketing inertia and away from those segmentation, message optimization, file maintenance/hygiene and tactical best-practices we all know about but which few practice on a consistent basis day in, day out.
Compared to direct mail and telemarketing initiatives, email is so much less expensive that there is a tendency not to conserve but rather to “blast,” not to segment but rather to “spray and pray”, not to groom the list but rather to grow it at almost any cost. At least that’s my theory and I’m sticking to it despite at least one obvious flaw. If it’s so relatively inexpensive and so immediately quantifiable, why do so few of us utilize email re-touches as a means of driving revenue cost effectively with minimal downside risk?
More emarketers are realizing email render, click through and conversion rates for their retouches that are 50% of the initial communication and more often than not, 75% to 100% of the original email’s performance. The 50% re-mail response rate typically seen in direct mail 30 days after the initial mailing is consistently being outperformed by email re-touches. This may be the result of a far shorter response window inherent in the digital space versus the direct mail space.
But regardless of the reason, going back to the most-responsive segment or segments of the file after the initial deployment using the same, identical message with a revised and relevant subject line is easily testable, immediate and quantifiable. It will enable you to look at not only comparative response rates but also make sure you’re not creating a new problem (like increased unsubscribe rates, spam complaints, etc.)
Obviously you’ll want to limit the re-touch efforts to those within your most responsive and larger sub segments who have neither opened nor clicked on the initial message. Doing so will minimize risks of over mailing and help hedge the performance bet. As an additional precaution, testing this approach as well as several different subject lines with a smaller representative sample is strongly advised and should not be overlooked.
Next time you’re celebrating a great email success, consider how to immediately leverage that success and the knowledge you’ve just gained into even greater revenues at a projected yet reliable efficiency rate. Being perceived as an efficient profit center is always a good thing especially during tough times.
Tweet This Post
Posted by George Palatine on March 9th, 2009
Email newsletter sign-up… triggered welcome message… hard bounced email. What just happened? The cause is most likely an email address input error by the user. Anyone who has been involved in email marketing has probably seen typos in their database, the question is what did you do to correct/prevent it?
Now this topic is not a new one in the email marketing best practices handbook. Unfortunately however even as validation technology improves the statistics have not. A 2008 FreshAddress study of leading internet retailer’s email validation practices reveals that email registration errors have shown little or no improvement from a similar 2006 study.
According to this research, 6.6% of email addresses are input incorrectly but suggests that it could be as high as 25% if no validation methods are used at all. Think about that, if your business collects 100,000 emails a year, 6,600 or more of them are wrong! Aside from the obvious economic implications of a lost opportunity there are additional concerns.
- Customer dissatisfaction with your company/brand because you did not perform as expected.
- Negative impact to your company’s digital reputation due to a significant amount of hard bounces in your list. This is a real problem because this will likely affect your email deliverability rates.
- Wasted marketing dollars due to bad list data.
Worst of all, just taking a few steps to insure accuracy could prevent this.
Many of these bad addresses are a result of simple syntax or misspelling errors. Some of the most frequent include;
- Using a comma instead of a period in your email address
- The @-sign mis-keyed as “2” (shift key not pressed)
- An accidental space within the email address
- Transposition of .com to .cmo
- Misspelling of Yahoo or Hotmail
So how do you fix these before they make it into your database? Here are a few simple/less technical suggestions.
- Make sure the size of the text entry box on your website form is large enough. The average email address is 26 characters long. If your form can only display 15 characters before the text scrolls out of view then customers could make a mistake and not even see it.
- Double entry of email address on a web form is a simple method but in most cases users will just cut and paste the first line into the second so it does not significantly improve your results.
- Reaffirm your offer and ask that people take a second look at their email address before they submit.
Some additional methods that require a bit more technical implementation but can be highly effective include;
- Email validation code on your web form. This can be a simple low-tech test for syntax errors to very sophisticated real-time validation/correction software. Which you use should depend on the value of each missed opportunity within your organization and your observed email failure rate.
- Confirmation page before posting the data to your database. For example, after a user clicks submit, a page can display asking them to take another look at what they just input. Be sure to allow them to make changes on that page or to confirm accuracy and continue.
The moral of the story is that you should monitor the email error rate within your organization and take the appropriate steps to reduce it. In this economy every little bit helps.
Tweet This Post
