There's Email, Again

Our focus on opens/clicks (when it comes to email) is so ridiculously misguided.

Email marketing serves many different purposes.

It should be used to teach customers about new product launches.

It should be used to sell winning items at above-average rates.

It should be used to replace expensive marketing tactics, especially among infrequent shoppers.

It should be used to convey loyalty benefits to the best customers.

And it should darn well be used to build a prospect list of shoppers who eventually become customers.

In the image below, we look by decile. Best new buyers are on the left, worst new buyers are on the right. Does a first purchase via email marketing align with the best new customers, or the worst new customers?

The top three deciles had about 20% to 32% of new buyers sourced via email marketing.

The bottom three deciles had about 2% to 3% of new buyers sourced via email marketing.

So yeah, email marketing matters ... if you build a list of prospects via email marketing you generate new customers who are much more likely to repurchase in the future than customers from other sources.

This is the point in the discussion where I get an email from somebody who says to me "Sure, this sounds good, but make it ACTIONABLE." 

The word "actionable", of course, means "I'm not doing anything until you give me a solution that works perfectly within my company culture and tech capabilities and I know you can't do that so that means I don't have to do anything." This happens repeatedly.

If you have a quality email marketing team (one not completely outsourced to vendors demanding that you create engaging content that increases opens and clicks), you have multiple email programs, don't you?

• One for Prospects.
• One for First-Time Buyers (i.e. a Welcome Program).
• One for Emerging Customers.
• One for Loyal Buyers.
• One for Lapsed Buyers.

Each program has their own cadence/frequency, their own messaging, their own creative look/feel, their own merchandising strategy, their own performance metrics.

This is where you tell me that this approach "isn't actionable".

And so we go ....