Friday, November 30, 2007

Analysis of Email Campaigns

Tracking Performance -

Measuring Useful Metrics

Tracking and analyzing the actions of your customers and prospects is critical to your success. After deploying several campaigns, you will have generated a mountain of response information that reveals invaluable data enabling you to create new and more effective approaches and offers. There are many software tools easily available that track, record and analyze all data pertaining to your email marketing campaign. Many sophisticated analysis tools predict future actions based on past behavior. Such tools are crucial for your campaign as response data is meaningless unless it can be analyzed and turned into valuable customer knowledge. Once you have the capability to track the vital statistics of your email marketing campaigns, the inevitable question becomes: how well are my mailings doing? Here are some guidelines on what information to gather and how to measure the performance of your ampaign from the gathered information. There are a number of different pieces of information that can be gathered when using any reasonably good email broadcasting service.

The five primary measurements are: the totals each of messages sent, message opens, click-throughs, bounces, and opt-out requests.

Total number of items sent must be accurately counted, based on reaching each individual email address only once. Opens measure the number of people who actually view the message using their email program.

I prefer to use “unique” opens, so that if a recipient views a message in their preview window, then opens it into a full size window, that this only counts as a single open instead of two opens. Clickthroughs are recipients that respond to your offer by clicking on a link in the email. Bounces are messages that are undeliverable to the ecipient. They could be “soft bounces” due to temporary issues such as a full mailbox or “hard bounces” from an invalid email account, but for our purposes here it means “people on your list who did not receive your message”. And last, opt-out requests are recipients who request to no longer receive email.

There are a wide range of results that can be measured for email campaigns, such as:
.. How accurate is the list (how many bounces out of total sent)
.. How active is the list (how many opens out of total sent)
.. How positive was the reaction to the offer itself (number of click-throughs out of total opens)
.. How negative was the reaction to the offer itself (number of opt-outs out of total opens)

The actual number of responses on any particular campaign can vary quite a bit. A newsletter whose primary job is to inform will not achieve the same click-through rate as a promotion, which is intended to get a specific response. The differing levels of permission within your list of recipients will also affect results. Until a list has been "cleaned" of bad addresses and those who are not interested, you may see far different data. Pruning these from your lists will help you improve your results considerably.

In order to account for these wide variations in factors, I suggest some general "baseline" ratios that should be achieved on any particular mailing. There should be more opens than bounces, or else the list is probably out of date. Also, there should be more click-throughs than opt-outs, otherwise the offer is poorly targeted or the list is of questionable origin.

In order to get the optimum response you will need to send two or three multiples of your email marketing campaign, each time using

a variation of the original offer. If they haven't responded by 3 attempts, it’s time to change your approach. The typical response pattern is that mailings 1 and 2 will have a similar response, with number 2 often having slightly fewer clickthroughs than number 1. Number 3 picks up the stragglers and undecided recipients, so the response will be much lower, but usually significant enough to justify the mailing. Please note that you shouldn’t necessarily just blast out three mailings one after another. For example, you might piggyback your first offer onto a monthly newsletter, send the second offer separately as a special promotional mailing two weeks later, then finish the series with the final offer in the next month’s newsletter.

It is useful to understand how the size of your lists is changing over time. By viewing how many people sign up for your lists each day, you can attempt to correlate list growth with other marketing activities that you may be conducting. It is also important to consider how many people are signing up for your lists versus how many are opting out of them. If your lists have been cleaned, and the overall list size is still shrinking, you need to reevaluate both your list acquisition strategy and the content relevancy of your mailings. Testing is critical to optimizing your email marketing campaigns. But in order to test, you have to measure first. Make sure you have a way to collect detailed information about your mailings, preferably in an automatic way. Careful analysis of the actual metrics will give you the information you need to take your email campaigns to the next level.

No comments: