Design and build requirements
It goes without saying that you need to be able to design and construct quality email in the first place, and that is hurdle number one for a lot of smaller companies using self service email packages – they don’t even get into the inbox.
Well crafted creative does cut through, and when combined with a policy of testing in different browsers, and scrutinising the soft bounce statistics we can build a reasonably clear picture of deliverability. We can even do on the fly split testing of creative and subject lines.
Good Data in Good Data out
But you should not overlook the importance of data. You get out what you put in, and with email marketing you would be well advised to have not just a name and email address but also a telephone number, mailing address, date of subscriber addition and a mailing address. This means that multiple marketing vehicles can be triggered from resultant data in your email lists such as DM and more qualified outbound phone calling.
Be Critical in interpreting email analytics
So what about the analytics you ask?
Well, most packages now give good data now on the following metrics;
Date and Time of opening
In addition you can use appended URL’s within the email to track users that click through into the web site as an individual marketing campaign.
Some email marketing systems, ours included can even track individual users through to a thank you page on a lead generation or ecommerce web site.
However, most people aren’t using the email reports properly. They send the emails out and tick the box, thinking that enough has been done.
With email marketing you have to take the time and invest it in understanding what interests your audience. If all you do is continue to send out uninteresting content then that’s how the audience will remain, uninterested. They will gradually unsubscribe and over time your list will become depleted.
Apply the Propensity Scoring Model to your email broadcasts
For each broadcast you make, you should use a method known as propensity scoring to understand whether the content worked for the subscriber.
We suggest the following propensity scoring methodology where 0 is cold and 5 is hot
5 – Subscriber opens email and clicks on links
4 – Subscriber opens email and clicks on single link
3 – Subscriber opens email
2 – Subscriber receives email but does not open
1 – Soft Bounce
0 – Hard Bounce
Most email marketing systems will remove the hard bounces automatically, and in general they won’t be any use to you hence the 0 propensity rating. However, a hard bounce might have become a customer, so don’t completely write them off particularly in a B2B environment, and if you have other data on the subscriber such as a phone number, don’t be afraid to give them a call.
Soft bounces need further investigation, as they may be due to a firewall, a full inbox or seen as SPAM. They aren’t a priority though and therefore have a propensity score of 1. We recommend you segment them into a new list and resend your email a few days later.
When a subscriber receives an email but doesn’t open it, it could be due to a number of factors such as date and time being wrong for the user, or that they have viewed it an a preview pane, or in their inbox without images allowed which hasn’t activated the tracking images within the email. Don’t necessarily make a judgement that they didn’t read the content. Therefore we score thsi group as a 2, and segment them into a separate list to measure success of future broadcasts.
The next prospect is the opener. With HTML email most email readers ask the user to download images. The very fact that the user has allowed the images to download, resulting in you tracking them in your statistics suggest a propensity to engage with your content. These users have a propensity score of 3, and should be segmented into a list for more targeted follow up emails.
Your statistics will then tell you how many of your subscribers have actually clicked on a link which shows a real affinity with your messaging. Segment this list now and create a more focused message relating to the content they clicked on. Send this out and then follow up with a telephone call if you have details on your database. Their propensity score of 4 means they may well be in research or buyer mode.
Subscribers clicking on multiple links are the highest propensity scores, and we give them a 5. These subscribers are clearly engaged by your email content and are worthy of personal treatment. If they’ve purchased before they may be receptive to a personalised discount offer, or if they are a B2B prospect it would be worth going for a sales meeting. Multiple clickers should be added to a high performance list segmented from other lists, so that even when they receive the same creative as others you can report on their interaction separately.
Being critical about data will make your email marketing campaigns more successful. Testing subject lines, personalisation vs non-personalisation. How do you continue the sales cycle to people who have engaged with your email?
Do you send some direct mail, make a phone call.?
Your email anlaytics will help you decide, but only if you take the time to analyse them.
About Roger Martin: Roger holds the Diploma in Digital Marketing from the IDM with credit, and has been working in digital marketing strategy for over 10 years across EMEA and the US.