At the risk of getting tomatoes tossed at me as the token marketer here, I took a contrary assessment of the MailChimp study that Dennis highlighted here. The study shows that emailing to an old file boosted complaints and unsubscribe requests, and depressed open and click through rates.
But that does’st tell the whole story. If I was a marketer only focused on response results, I’d look at this data another way. I’d say, “Hey – I got 7,688 clicks on the “bad” list and only 6,925 clicks on the “good” list. I doubled my bounty! So mailing the bad list was totally worth it! Let’s do that again. And again! Party on!”
Many marketers sstruggle with this sort of situation. We know the best practice is to not email old files. We know the best practice is to refrain from emailing that extra campaign this week. However, at the same time we are under pressure to drive short term revenue. “The numbers are down, just send another email campaign.” And it works – the extra email campaign drives revenue.
In this case, not following best practices generated a lot of traffic. The MailChimp study doesn’t say if the old file clicks converted at the same rate, but let’s assume that they were at least comparable. There was some gold in that there old file. Of course there was a lot of shiny rocks, too.
So, if you are going to break the rules, be smart about it and make sure you have enough data to make informed business decisions. You need to know if the risk of complaints or brand degradation from annoying your subscribers outweighs the short term revenue. Consider these protections and optimization strategies.
1. Make sure you mail from a distinct IP address, and quarantine the data to be sure it’s spam-trap free.
2. MailChimp could never assess the value of this old file if they didn’t have the deliverability data. You must seed these mailings if you want to make good decisions about future mailings. Deliverability is a marathon not a sprint. Track as many campaigns as you can.
3. Finding the gold in this file meant mailing the entire file. But next time, consider only emailing those who opened. Or be sure to remove records that have not responded in 12 months. Keep the old file risk to a minimum.
4. Track the results all the way through to the bottom line. In this example, if the clicks from the old file didn’t return actual purchases/conversions, then they are worth a lot less. So when you do your ROI, be sure to use the numbers that are most meaningful for your business.
Although sometimes we feel like marketing strategy is equivalent to holding up a torch and peering in the dark, murky back room of a cave, in email marketing we have enough data to make good business decisions. Shame on us if we don’t take full advantage.
Last 5 posts by Stephanie Miller
- Are You Ready for the Ultra Managed Inbox? - August 30th, 2010
- Privacy Updates & Regulation Ahead for Digital Marketers - June 4th, 2010
- Half of Global Email Users Knowingly Click on Spam - March 25th, 2010
- The Importance of Inbox Placement Data - February 14th, 2010
- What Will It Take To Get You to Authenticate - August 4th, 2009
Hi Stephanie,
You're 100% correct about the "marketer focused solely on results" thing.
As I was putting together the stats for the "bad" list, I thought to myself, "Oh crud—why couldn't they have gotten fewer clicks?"
Because, like you said, over 7k clicks isn't something to laugh at. No marketer would throw away 7k clicks.
But, as you recommend in tip #1, the least they can do is not allow those old/bad email addys to "taint" their new/active email addys. Quarantine them and run tests to find the gems in them.
No sense allowing old emails to bring down your IP's reputation.
Ideally, we'd add a seedlist to the "old" list, and we'd add a seedlist to the "actives" list, then we'd send campaigns. Then, we'd send another campaign to the combined (old+new) lists, and add a seedlist to that one.
Then, compare results to see if including old emails in your list brings down your overall deliverability.
(Un)fortunately, no client in their right mind would allow us to experiment with their list like that.
Thanks, Ben! Actually, what you suggest seems imminently possible if you quarantine the two lists on two separate IP addresses, right?
And, the key unanswered question, although you may not be able to say, is the conversion rates off the clicks. If the clicks generated by the old file convert at a much lower rate, then our argument is complete!
THANKS for the great study and for your comments.
Having done something like this with some lists at the last place that I worked at, what I saw was that conversions on the older lists were much lower and much more likely to be blocked.
What we found to be more effective, in terms of overall delivery and managing complaints was to establish a good list with the newer data and gradually add in older data. With a bit of math, you can figure out how much older email you can add in without going outside of acceptable bounds.
That said, we were rather cautious with the older addresses that we added in, restricting them to once-engaged subscribers and a few other restrictions, based on the individual customer needs.
is extremely helpful resources to guard our time.In case you under no circumstances pay consideration to your self ,please grasp this probability.a couple of days ago,I purchased .