• by editor , David Baker , July 21, 2008

Tuesday, Wednesday, Thursday? When is the right time to send an email? That may be the right question, but is the answer what you are really looking for? This is probably the most widely asked question that email marketers will face. People used to be fascinated with time of day and day of the week; now that we are well over 15 years into email marketing, I see less credence in these industry surveys of 1,200 respondents that try to help you answer this broadly.

There are several ways to help answer this question.

Leveraging behavioral insight and customer segmentation is a must. When consumers opened the email, the delay between view, and some form of response (site visit, click-through, registration, order) are all valuable data sources -- but if you can't get such factors as site visit depth, duration on site, how many visits to site from first email to last, and isolate them to a subset of high-value and high-invest segments, it's meaningless in mass.

Over the years, we've tried about every tandem of cadence and timing, and I've found one thing that has remained constant: consumers are easily conditioned to good things. I've always been a big proponent of keeping newsworthy or community-centric content to a consistent cadence and timing. The email inbox is not like the mailbox. We know when the electric bill or People magazine will arrive. Consumers do expect certain types of email to be there at certain times, but the best day of the week is an amazingly dynamic thing that continually changes, and optimizing it can be elusive.

I have a scenario for you; see if your opinion of this question changes.

You send an email promotion on a Wednesday at 7 a.m. EST, and 95% of your respondents interact within 24 hours. This day has historically trended higher than any other day of the week (in terms of response and number of orders). After testing the difference between a Wednesday and Saturday, you find that while you have a 20% difference in performance between the two days, 30% of two of the highest value customer segments buy on Saturday rather than mid-week. You also see a consistent trend in AOL that trends higher on Saturday -- and within that segment, you have a higher average order value and a larger concentration of high-value customer segments. While the aggregate campaign performance proves you are generating more business results by sending on Wednesday, you may have more traction with high-value segments on the weekend. So, what do you do?

-You could personalize the delivery of this weekly promotion to the purchasing trend of the consumer. - You could send a promotion on both Wednesday and Saturday and hope to maximize both worlds ideally.

What if this changes by time of season or nature of promotion or some influence from your retail front (increased media promotion, new keyword buys, television promotion, POS promotion)? Do you factor these influences into your story?

The point of this column is not to tell you when to send email. It's to point you in the right direction toward applying a more accurate view of measurement, what influences can impact this behavior and the results of your email program.

Ideally, email should do several key things with every program:

1. Drive increased performance, which can be defined many ways financially (increased sales, increased profitability). 2. Influence a condition (increase purchase frequency, increase average order value which can translate to increased customer loyalty). 3. Build a credible story behind your results that is translated by key customer segments behavioral changes (customer segment A has this trend, and these factors influence purchase, and customer segment B relies more heavily on the site to support purchase decisions).

There are a million stories we can tell with each campaign, but only a few that stick. Build your hypothesis and results to last.