When an E-mail Marketing Campaign Goes Bad

It’s happened to all of us at some point. You’re marketing a new product or service for the first time. You use your best instincts and knowledge to develop a campaign strategy, identify targeted email lists, set the offer, write the copy, and send. Despite your best efforts, the response is less than you anticipated. I can’t tell you how to avoid the situation (wish I could, because then I’d really be in demand as a consultant!), but I can provide some tips for damage control both before and after the send.

A quick note before we start. In this column, I focus on CTR, but the real criterion of campaign effectiveness is determined by your end goal, not CTR. In the example below, our goal was to get people to opt in to a free email newsletter. The number of new subscriptions didn’t match our goal. That’s the point at which we went back and focused on the campaign metrics. If we’d had a different goal, say selling enough email newsletter subscriptions to cover our costs, we might have made it with the same CTR. Your goal, not CTR, is your benchmark for success or failure.

Before the Send

1. Don’t be blinded by cost. With lists, you get what you pay for. For a recent mailing, I recommended a couple of lists to my clients, based on list research I conducted and provided them. They came back with a decision to mail to one of the lists I recommended and two I had rejected. The lists they chose were significantly less expensive than the ones I selected. They felt they could mail to more names for a lower cost.

   Here are the results:

   List	List Cost ($ CPM)	CTR (%)	Effective CPC ($)
   A	350	1.72	21.88
   B	100	0.32	35.71
   C	100	0.18	62.50

   In the end, those less expensive lists didn’t perform nearly as well. They ended up costing a lot more on a per-click basis. You never know going in exactly how things will turn out, but what appears to be a “bargain” list often isn’t.

2. Mail to more than one list. If you mail the same creative to multiple lists and track them individually, you’ll gain more insight and potentially more leverage in negotiating a make good (see below for more on make goods).

   In the example above, the same creative and offer were mailed to all three lists. Two performed very poorly; the third list, while not a superstar, returned an acceptable CTR. As a result, we were able to theorize the creative wasn’t the problem, those two lists were. If we’d mailed to only one list and it bombed, we wouldn’t have that key learning. Ditto if we’d sent significantly different creative to each list.

3. Break out tracking and reporting by list. This applies to open rate as well as CTR and holds just as true offline as online. You know the adage, “I know half my advertising is wasted, I just don’t know which half”? This provides the information you need to identify which half (or which two-thirds).

After the Send

1. Talk to your list broker. Ask your list broker to review results with you. Don’t be shy about requesting a make good if you can make a case that the lists underperformed. A make good is an additional mailing at no cost to a similar audience and quantity to make up for poor list performance. Many brokers will offer this without you having to ask, if performance is really dismal.

   In the example above, the broker agreed to a make-good send for the low-CTR lists, using hotlines (the most recently acquired email addresses) from similar lists. The mailing used the same creative as the original send, was equal in quantity to the two original test lists, and will be sent this week (I’ll tell you how it performs in the next column).

2. Know what a make good will — and won’t — achieve. Make goods are always worth doing. Depending on how they’re done, they may or may not meet all your original send’s goals. For instance, a hotline send such as we’re doing for my client will meet short-term goals but won’t help much with long-term goals.

   The short-term goal is to triage to generate a higher final return on investment (or, more precisely, a lower cost per new subscriber, as this is a newsletter acquisition campaign).

   The longer-term goal is mailing test quantities of significant sample size and identifying lists we can roll out on in the future. There won’t be enough emails from any one list to know it will perform if we roll out. We may be able to rent a similar mix of hotline names on a regular basis and have success with that, but that’s not where we started. In that scenario, the potential universe of names is smaller.

3. Marketing is an ongoing learning process. You can learn as much from a campaign that doesn’t work as from one that does. Keep a record of what creative was used, what lists you mailed to, and the metrics. Build on what you’ve done. We now have a baseline response rate for the one list that didn’t tank. We consider that our control and will use it as a benchmark for any future campaigns.

Hope this helps! I’ll tell you next month how that make good goes.

Jeanne

Meet Jeanne at ClickZ E-Mail Strategies in New York City on May 19 and 20.