What’s the right tone when communicating with your winery’s email database?

Hi everyone! My name is Simon Garlick and I am chief geek and wordsmith at a small Adelaide marketing agency. I create and manage Internet marketing campaigns for a number of South Australian businesses; I’m proud to count among their number some Barossa wineries, and it was thanks to those relationships that I attended the BGWA’s Barossa New Media Summit last Tuesday (if you were there, I’m the guy up the front who turned around and asked “who is on Twitter right now?”)

During the audience-participation segment of the summit one question really stood out to me, a question about the best tone to use when communicating with a recipient database. I was discussing it with James after the summit and he suggested I put my thoughts together for Barossa Dirt.

What is the best tone to use when communicating with a recipient database?

Well, it depends. Every organisation is different; every recipient database is different; every individual communication is different. There is no one-size-fits-all “best tone”. But here’s what an online database-driven marketing campaign can do: it can reveal the best tone for the task at hand, and it can do so more accurately than traditional marketing methods, in a fraction of the time and at a fraction of the cost.

What do we need to do?

So imagine we’ve got two messages with distinct and differing tones. How do we find out which is the better of the two?

Here’s the basic method:

1. Identify a criterion by which success will be measured.
2. Try both messages.
3. Compare the results against the success criterion from (1).
4. Whichever version performed the best against your criterion for success is the winner. Use it!

I know, it looks obvious right? No gut feelings, no intuition, no fluffy marketing talk. One of the two options will have worked better than the other, therefore that one is the best. That’s all there is to it. But doing it with modern electronic tools means we can automate the process, saving lots of time and effort.

Let’s plug the question at hand into our test steps above and see how it works for the imaginary Barossa Dirt Wines.

First up we have to identify what we’re going to measure in order to determine success. Here’s one variable that can indicate the effectiveness of our tone: the number of people who actually open the email. Since that’s our yardstick, the variation we test will be something that influences whether or not recipients open the message: the subject line.

Next up, let’s create two email messages for Barossa Dirt Wines, two messages identical in every respect except for the subject line. Version A will be personal and chatty (“G’day <firstname>! You won’t believe this month’s news – read on for more!”) and Version B will be businesslike and formal (“Welcome to the Barossa Dirt wines email newsletter August 2011″).

We’ll run the test like this. Imagine Barossa Dirt Wines has a database of 10000 recipients. We’ll send Version A to a sample – say, 10% of the database, or 1000 random recipients – and Version B to another 1000 recipients.

We’ll send the email designed in such a way that we get “pinged” each time the email is opened by a recipient. Note: this is entirely invisible to the recipient; we’re not talking clumsy “read receipts” or anything.) We will record those “pings” over a timeframe long enough to give us some useful data – say, 2 hours – and then whichever version has been opened by the most recipients after 2 hours is the winner and gets sent to the remaining 8000 recipients automatically.

Running the test

So 1000 recipients find a cheery greeting like this sitting in their inboxes:

and another 1000 find this instead:

At the end of the two hour test period we check the numbers and compare open rates…

We have an answer

The email that most effectively catches the attention of the recipients is Version A, the more chatty of the two, improving on Version B by about five and a half percent. Five and a half percent may not sound like much but remember we’re talking about a database of thousands of people. With nothing more than a change to one line of text we get the attention of hundreds of potential customers who otherwise would have ignored us.

I wonder how much wine those hundreds of people might buy from our imaginary winery because we successfully got their attention. I wonder how much wine they would buy over the entire lifetime of their relationship with us.

It adds up. It adds up to a return on investment (ROI) for email marketing that blows any other sort of direct marketing away.

Now you see why I love this stuff!

But wait, there’s more

“Hold on,” I hear you say, “surely the effectiveness of the email depends on a lot more than just the subject line?”

You’re right of course. Many things can influence the effectiveness of your email – copy, layout, imagery. And how do you know what works? You create alternatives and test them!

Let’s imagine once again that we’re sending out an email for Barossa Dirt Wines. We’ve settled the question of tone for the subject – now what about a call to action? More specifically, will an invitation to visit the Barossa Dirt Wines online store work better at the top of the message or immediately after the first article?

The process is the same. We create Version A with a “click here to visit our online store” call to action at the top:

Then we create Version B with the call to action following the first article:

Again we set sample sizes for the test, a time period, then send both messages and record the “pings” we get back – this time specifically tracking clicks to the online store instead of message openings.

The test isn’t over yet, but I think Version A is going to be the winner. At the conclusion of the test Version A will be sent to the remaining recipients, and by the looks of it having the online store link at the top of the email will directly result in hundreds if not thousands of extra clicks on the link to the Barossa Dirt Wines online store. Think about it: moving a line of text by ten centimetres on screen directly results in hundreds more people visiting the store. 

The minutes of work it took to set up the variations and run the test have more than paid for themselves in terms of visitor numbers, and over the lifetime of the relationship we just maintained imagine what those store visits would add up to in terms of revenue.

Could you imagine doing this with a paper mailout stuffed into stamped envelopes? It would be a nightmare. The time and expense would be huge and any information obtained would be nearly useless.

Or to look at it more positively: the time and expense saved by communicating intelligently with your database via electronic means instead is huge and the information obtained is invaluable.

This is why I am such an advocate of email marketing done well. It’s a marketing method that’s perfect for the wine industry, as wineries usually possess customer databases already thanks to cellar-door and mail-order lists. And let’s not forget that while the tools are new the goal is unchanged: maintaining relationships and serving the customers better.

OK, that’s enough

Well that’s me for today. If you made it this far, thanks for your attention – I hope this has been useful, or at the very least enlightening as to how and why we do this stuff.

I’m happy to answer further questions about these sorts of online activities, so please feel free to leave comments below. Alternatively I can be contacted via email or by phone at 08 7127 0435. Thanks once again to the Barossa Dirt team for the opportunity to share this with you all; to James, Dave, and Andre for their discussion; and to the gentleman – whose name I never caught – who asked the question that got me going with this.

+Simon Garlick

Tiny URL for this post: http://tinyurl.com/3hx66yz