Always Best to Try a Test - Testing to Improve Marketing Results

By: Bryan St. Amant


One of the best steps you can take to ensure great results from your email marketing is to test various components of your campaign before mailing to your entire list.  By taking the time to learn which offers, headlines, and delivery style your customers like most, you can boost response rates -- and your sales -- by 25% to 100% with every mailing you send.

And the good news is that with email's low cost, lightning-fast customer response, and easy-to-measure results, there is no other marketing medium that's easier or more profitable to optimize with proper testing.

That's why this month's wine marketing tip is:  It's Always Best to Try a Test.

Let's look at an example.  Suppose you're marketing a winery event.  You develop an email subject line that reads "Free Event for Wine Lovers."  Your sales manager suggests the intrigue of "Your invitation is enclosed"  And your event manager thinks "Exclusive Winemaker's Dinner" is just the ticket to pull your customers through the door.  If you want the best results, which subject line do you choose?

You could always guess, but you'd only have a 1-in-3 chance of choosing the winner.  And over time, you'd be settling for inferior results from the majority of your marketing campaigns. 

Our advice is to test all three subject lines on a portion of your list, and let your customers identify the winner with their response.  Then when you mail to the rest of your list, you can be guaranteed you're producing the very best results -- with no guesswork required.

Setting up a test can be as simple or as complex you prefer.  If you work for a winery with hundreds of thousands of names and a department full of marketing statisticians, you might want to run a multi-panel, multi-variable test matrix.  But for the rest of us, a simple A/B comparison is usually all it takes to reap the benefits of testing. 

Here's how a simple A/B test might look for our example:

  • Free Event for Wine Lovers 22% open/7%  conversion
  • Exclusive Winemaker's Dinner 19% open/9% conversion
  • Your Invitation Enclosed... 31% open/5% conversion

After three tests to 500 names each, we can see that the "mystery invitation" was effective in getting people to open the mail, but once they learned the details of the offer, fewer customers actually 'converted', or signed-up for the event.  Our "Free Event" subject line did a good job on both counts, but "Exclusive Winemaker's Dinner" was the clear winner, generating 40% more response than the runner-up and 80% better results than the loser. 

And while these headlines may be artificial, the results are not.  A 50% to 100% spread from winner to loser is not uncommon when testing the key components of your email campaigns.  So even if you guess correctly much of the time, testing will still greatly improve your long term success.

So what components should you test?  The subject line is a great place to start, but for best results you should be testing all of the elements which could significantly impact the success of your campaign. Our suggestions are: List, Offer & Creative.

If you have a choice of which list(s) to use, by all means test your lists first.  The list you use will drive results more than any other element of your campaign. 

But suppose you only have one list to mailing list (i.e. your own customers or prospects).  In this case, you can still focus on optimizing your offer and creative.  For example, which offer will pull more, a 10% discount or a special bottle signed by the wine maker?  Don't know...then test!   Does "pretty" HTML email outperform simple text-based messages to your customers?  Don't know...then test!

Other variables you might want to test are long copy vs. short copy, the timing of your emails, and especially the design of the landing page your customers see when they click on your offer.

Of course, there are a few basic rules you'll want to follow to produce useful results.  To interpret your results clearly, you should test one only variable at a time.  To track your results accurately, you'll want to assign unique keycodes or URLs to each of your test mailings.  The test names you mail to should be selected randomly from your overall list, and the size of your test mailing should be large enough to generate a meaningful number of conversions.  

But like tending your vines out in the field, a little effort will go a long way to produce the best results possible from your email campaigns.

So this year, while your vineyard manager and wine maker are working their special magic, why not try a bit of your own by testing and optimizing your email campaigns?  You'll be thankful for their results, and they'll be thankful for yours.