You are using an old browser that may not function as expected. For a better, safer browsing experience, please upgrade your browser.

Log in Become a Member

If you want to reach your destination by car, you (hopefully) look ahead; correct if you go too far to the right or into oncoming traffic, turn in the right places and brake if unexpected obstacles appear in front of you.

It goes without saying - so why am I writing this? Because this is exactly the approach that is far too rarely taken for granted in wine marketing via email. There - to stay with the image - people just step on the gas and their eyes remain closed. Yes, and in the end, many then ask themselves why they don't reach their goal or why it comes to a crash.

However, unlike driving a car, in email marketing we need specific tools and metrics to truly see. I'd like to talk about some very basic metrics today:

Delivery Rate

So, for example, if you send out 1,100 emails, 1,000 of them will arrive. The other 100 end up in spam filters or were sent to email addresses that are no longer valid. This is called a 1,000/1,100 delivery rate, or 91%.

Measuring the delivery rate is very complex. So let's not even try. However, we do of course know how many emails come back as so-called bounces. These are the well-known and often somewhat cryptic replies from the mail server that a mail cannot be delivered for certain reasons. The number of bounces should be well below one percent for well-maintained recipient lists. This is called the bounce rate.

Open rate

Back to our example. If 300 of the 1,000 delivered emails are opened by the recipient, we are talking about an open rate of 300/1,000 = 30%. The open rate is one of the most important parameters for every email dispatch. It is measured by determining how often one of the images in the email (possibly just a very small and invisible "pixel") is loaded from the server. The images are only loaded by the e-mail program when the mail is actually opened. However, there are a lot of sources of error here: Not every email program loads images automatically, some providers work with a cache and load the image only once for all their recipients, etc. Therefore, the measurement of the open rate is always inaccurate and actually does not really reflect the number of emails opened. That's a shame and unavoidable. Still, not really a problem. More on this below.

Normal measured open rates hover around 20% to 40%.

Click-through rate

It is already an important goal if your emails are delivered (=delivery rate) and read (=open rate). However, it is usually also intended that the recipient clicks on one or the other link. In other words, to inform themselves further or to buy something. Therefore, it is also important to determine how often readers have clicked on a link in the newsletter. This frequency can be expressed by the click rate. This tells us how often the recipients of the emails have clicked on a link.

But in any case, the click-through rate is a percentage as follows: Clicks / Number of Emails. Both numerator (clicks) and denominator (number of emails) are not clearly defined. For example, the denominator is sometimes the number of emails sent and sometimes the number of emails delivered. It is also conceivable to use the number of opened emails. Also the counter can be calculated gross (multiple clicks of a recipient are also counted) or net (each recipient counts only once).

For you, the exact formula is not that important either. Just take the one that your tool determines.

The click-through rate depends very much on the topic and the email. For example, the click-through rate for an event announcement, where everything is actually already said in the email itself, is also less than one percent. However, well-optimized emails with strong click animation can reach 100% or more (*).

Very accurate interpretation despite measurement errors and inaccurate definitions

By now you may be wondering why, despite the errors and unclear definitions described above, the metrics presented are so important. Because, in fact, it is not the absolute number of delivered or opened mails or clicks that matters, but only the change in those numbers. In fact, as long as the measurement method and the formula used remain the same between mailings, the errors in the comparison between different mailings cancel each other out! (**)

And that brings us to the core task of any marketer:

Testing, i.e. trial and error

If you send a regular newsletter to all your wine customers, you want to permanently increase the effectiveness of this newsletter. That's exactly what the three metrics are for. Learn what recipients care about by watching the metrics. Like an artist who learns from the applause of his audience what goes down and what doesn't. And of course, the next time he leaves out the things for which he doesn't get applause and strengthens the things where the applause was particularly loud.

Unlike the artist, you even have the option of splitting your audience during the same "performance" and trying out several variations. This is then called split testing in marketing. To do this, use a random (!) criterion to split your recipients. For example, send one variant of your newsletter to all customers with even customer numbers and the other variant to your customers with odd customer numbers. Then compare the above characteristic values for both variants. Of course, these should only differ in one place. For example, by a different subject line, different layout, different sender information, and so on. You will then see exactly which variant you should use in the future. In this way, you will improve your mailings one by one! And your customers will be happy too, because they will be addressed more and more relevantly and better by you in the future!

All well and good. But how can I measure all this?

Basically, I recommend, especially to my probably less technically savvy readers, to use a service provider for their email marketing. The cost of this is kept in check, and in return you have very good statistics and tools. Bounce management (sorting out delivery errors), subscription and unsubscription, etc. are well and professionally solved here. These service providers also ensure that their sending servers do not end up on spam lists and thus a high delivery rate is achieved. You can find these service providers in Google with the search terms "email marketing service provider newsletter" or similar.

Of course, there are also software solutions for your own PC, with which you can do all this yourself. Please do not underestimate the effort and possible sources of error. (***)

Final note

I would be pleased if I could create some clarity about the importance and interpretation of the most important key figures with this contribution. I have explained many online marketing terms in the wine marketing glossary.

As always, I welcome your feedback and questions.

(*) A little excursion into the wonders of marketing mathematics: Don't be surprised about click-through rates even above 100%. On the one hand, this is because individual users can click on multiple links in the mail. And if, for example, the number of measured opened mails is used as the denominator, the actual number of opened mails can also be higher, precisely because of the inadequacies described above.
(**) You don't care how many decibels your spouse speaks to you and how this is measured exactly. You do, however, know how to appreciate it when it gets louder!
(***) By the way, at Wein-Plus we work with the Supermailer. We have supplemented this with our own solutions and can thus send very professional mailings that are relevant to the recipient. Our database and the presented parameters help us to make our emails as relevant as possible for the recipient.