Digital Marketing Insights
June 07, 2012 | By Justin Williams
Here's an article I wrote for iMedia Connection:
I'll be the first to admit that when I receive revenue stats like "95 cents per email sent," an enterprising part of me thinks, "If I send more email, I'll make more money!"
The good news: This could be true. The bad news: Sending more email could be destructive to the precious reputation you've built -- and to your revenue (even if it takes a while to take effect).
So do we give up that extra revenue for fear of losing it all? Do we simply ignore those of our customers who are willing to convert more frequently to appease those who don't want more email? Do we gamble with our revenue?
With today's technology, absolutely not.
Look at cadence as an opportunity to be even more respectful (i.e., responsive) to each customer's desires. Begin to learn about how many emails your customers want and when they want emails (and what they want in those emails).
This is harder than it sounds. Your list is not populated by uniform, static automatons who respond with machine efficiency to your marketing efforts. Each subscriber has different needs during different seasons and during different stages in their life. And these needs are different than the next subscriber on the list.
The sophisticated email marketer should be able to vary cadence -- frequency and timing -- based on users' preferences. The sophisticated email marketer can also find at least three ways (maybe more) to discern these preferences and assign different subscribers to different lists.
Method 1: Ask them directly
As part of an enhanced welcome program (or whenever you choose), poll your subscribers on how they'd like to receive emails. Start with just one parameter, such as the stage of their life. For example, you could send a poll asking:
Which best describes your current living situation?
a) Renting an apartment
b) Looking to buy a house
c) Own a house
d) Transitioning, looking for a new place to live
e) Wandering or homeless at the moment
If you were a home retailer, you could use this information to alter frequency and timing appropriately (not to mention content). Test it with 20 percent of your list, and see if you get a lift in your KPIs over standard promotional campaigns.
Method 2: Observe and assume
Separate your list into a few groups and experiment. For example, if you currently mail all of your list three to four times a month, make three groups. For group A, do what you've always done. For group B, send just two emails a month. For Group C, send six times a month.
Alter your strategy based on your intuitive knowledge of your lists, but at least try more frequently and less frequently (if you're testing for frequency only). You can do more groups if you like, and you don't have to put an equal amount into each group (e.g., five test groups could have only 10 percent of your list if you're worried).
See what happens. Really get into the data and see if certain subscribers opened more or converted more when they got less email. Assign them to get less emails moving forward. If some subscribers didn't open more despite receiving more emails, they shouldn't continue to receive more frequently.
Remember to control your tests for one variable. You shouldn't be testing new content in one of the groups that is also receiving email more frequently.
Also be sure to give these tests enough time to run. In the example above, I'd say a month would be sufficient, depending on the frequency. You should probably get 10 emails out there in your control group to make good inferences.
Method 3: Look and guess
If you're having deliverability issues, or for whatever reason you can't wait to hear back from your subscribers on what they'd like to see cadence-wise, all is not lost.
Build some segments from the data you already have. Build a group of subscribers that open 25 percent or less of the emails you send. See what happens if you email that group half as frequently.
Build a group of subscribers that convert often. Try to email them more often, and see if that increases. Take small steps, and be sure to measure the difference.
Ideally, you'll identify a group (e.g., subscribers who open 25 percent or less of the emails sent in the last year) and, from that group, build a control group and a test group that will get the adjusted frequency. See how the stats compare.
This method is not as good as the first two, but it can help.
Don't treat everyone the same
Your subscribers want to receive a different amount of email from you at different times in their lives and at different times of the year. The more responsive your program is to those desires, the more relevant you become, and the better your campaigns will perform.
If you're wondering whether you should send email less or more frequently, test it. There's a good chance your email service provider offers the functionality, and perhaps even the services, to help get you started or manage the process entirely.