Request an Esitmate
Send Us an Email
Being Information Driven: Understanding the Meaning of Data
October 12, 2010
There are many web statistics applications that claim to tell you things that you can’t actually know, you can read more about that in our web statistics article in the side bar. As it turns out, email and eblast metrics are not free from the detail myth either. Many providers try to tell you things that they simply can’t know.
What’s more, it’s important to take a broader look at how email marketing can affect your overall traffic:
1. For starters, it’s not just about opens and click-throughs. Some users will see the email in their inbox and simply fire up a web browser to see for themselves what’s new on your site. We see this all the time in looking at our client’s data. The actual traffic to their websites that specifically targets the content highlighted in their email campaigns will shoot through the roof, sometimes doubling or tripling their actual click throughs from the eblast itself.
2. Unlike social media, eblasts tend to live a lot longer than their social media cousin tweets and posts. An email could remain in someone’s inbox for weeks before they actually find the time to read and click. This is one of email’s strengths, they can be kept by your users and then be put to use when their schedule permits.
So what can you know? Like web statistics, there are many items that can be tracked. Depending on your situation, some will have more value for you than others. Here are some fundamental metrics and their basic meaning:
This ought to be a no brainer, but in many organizations you have to import and export your data between tools, so your actual number sent, may not match your actual mailing list. With GeekPAK, your mailing list is always current with no need to import or exchange data so you always have a reliable number of emails you sent. Because GeekPAK automates this process for you everytime a user subscribes, your list could grow the second you hit the send button, and you didn’t have to do a thing.
This is one of the best examples of data overshadowing information. Don’t get distracted by numbers like this one. Focus instead on what you want users to do. Outcomes mean more than a metric that, well isn’t really a metric.
Open rates are at best a guess. Typically, special images are used to track each open, but not every email client automatically downloads images so the count can never be accurate. What’s more, a given person could open the same email on their work computer, their home computer, their iphone and their ipad...so how many opens is that? Does one person opening the message 4 times mean more than one person going to your website and reading 10 pages?
There are many other examples we could give about why open rates are suspect, but the punch line here is that these numbers like web stats can have meaning longitudinally, but never take these numbers to mean precisely how they read.
Now we’re getting somewhere. Click rates track whether or not a person actually clicked an item in your email and went to your website or took some specific action. This is one of your most important metrics and can tell you a lot about what your users value most about your content.
As it’s name suggests, this tracks how many times your email was forwarded to a friend. Again, this is a fuzzy metric because not everyone is going to click your “forward to a friend” link. Some people, rather many people will click their email client’s forward button simply out of shear muscle memory. In fact, reflecting on user behavior shows how all of your metrics are on the fuzzy side and can never mean exactly what they say.
In addition, we think of page views on your website over the course of a week or so following your eblast to also be a kind of forward. Some users will jump from their mail client straight to their browsers and explore on their own. Web traffic always spikes around eblasts, so we think of this as an additional kind of forwarding of visitors to your website.
Keeping an eye on how many people are unsubscribing is critical. If people are leaving in droves you obviously have a problem. Keeping an eye on this can tell you if you are eblasting to often, sending information your users don’t find valuable etc. In short, unsubscribes are the opposite of results and are in many ways as important a metric to stay on top of.
Not every organization has a widget for sale, or a letter they want their visitors to send to their congressman, so identifying a specific goal to track can be difficult. This item more than any other should be addressed before you send your eblast, and even more importantly, not misinterpreted after you send. Correlation is not causation, so if your eblast doesn’t have an extremely obvious outcome like purchasing a specific product, try to take a longer term view at your outcomes. Study your trends over time to see what impact your eblasting efforts are having over time.
One of the key things about eblast metrics is spending time thinking about what you want to know from your statistics, and what results you hope to get. It’s not enough to just think about the data points, what do you actually want your users to do? Whether it’s a sale, or a particular action, chances are, clicking on a link isn’t your end game. Data begins to become information when you can match it to your goals and outcomes.
What’s more, because web user behavior goes beyond the email client, we believe looking at web traffic in conjunction with eblast metrics can also provide valuable insight into the effectiveness of your campaign, as well as what your users are actually interested in reading. They may find you interesting, and they may crawl all over your website, without ever clicking a link in the eblast you send.
With the wider view you can start to notice what your users want to read about and use that to make your eblasts more successful by responding to your user patterns. What’s more at GeekPAK because our eblast tools integrate with our CRM, you can actually target specific users with specific interests with every eblast. Overlaying web traffic, user behavior, and CRM profiles transforms you from being interested in data to being driven by user information.