In the old school days it used to be you’d measure an email campaign’s worth by the number of opens it produced, and the number of clicks to drive traffic to your website. But in these more challenging days, with the open metric being horribly skewed by Apple’s MPP among others, are just clicks enough anymore? Most email service providers should be able to provide conversion tracking, and this by far could be the most important metric – if the purpose of your email is to sell things, then just how much have you sold via your email?
A conversion of course doesn’t just have to be for retail success – you may want to track people that showed an interest in an event or your services, or clicked through on to a specific call-to-action – whatever it may be, it’s all trackable. Depending on the ESP, it could require a little bit of extra code on your website but here’s an idea of how easy it would be with us were we your ESP:
Step 1: Capture your user
We generate a unique conversion ID for the user and campaign which will get included in any URL back to your website from your email. You just need to make sure you can capture the value of the parameter, e.g. cid (this is the default, but you can change it to something else if it unluckily clashes with another parameter of the same name), from the URL and store it in a cookie or session variable to be used later on the conversion page. So, when your email gets sent out, your URL will end up looking like this and you can just snaffle up the cid:
All you need to do is stick a beacon, or an invisible 1×1 pixel sized image, onto the page after a conversion has been made, e.g. the “Thanks your order has been completed” page if you were tracking sales. Then you can drop the captured ID into the src link for this image back into the platform that will record it for your specific campaign.
You can also juice it up a bit by adding the amount they spent, and even labels to help you break the conversions into categories to see if one area is doing better than another e.g. if you sell white good appliances you could stick in what type of sale you made, a WashingMachine or DishWasher or MultipleProducts:
We all know that despite our best efforts, things can go wrong, normally in the form of a page refresh. In some cases it may not matter, but if you are tracking physical sales you definitely don’t want those being inflated. To prevent that happening you can add something unique to your link, for example an order number, and this means your conversion can never be counted twice.
Within the platform, all you have to do is toggle it on:
The links from your template will now include the cid, or whatever you want to call it, and you should see your conversions begin flooding in. As you can see here, you can also add an extra follow up to a conversion if you wanted to, e.g. trigger an email to them with extra information automatically which opens up a whole new world of possibilities.
Instant analysis at your fingertips
A lot of this could be tracked in things like Google Analytics to be analysed later, but the advantage of doing it within the campaign itself is you then have visibility of your most active customers, so not only do you instantly see how well your campaign performed, you can then build up a better profile on your users to target them further.
Why not get in touch to see if we can help enhance your campaign analysis?
Are those pesky email applications messing with your design? You didn’t want that address to be automatically linked to Maps, and you certainly never asked for telephone numbers to be underlined! It’s time to squash the bugs.
The battle begins
Overriding a piece of email software’s functionality often isn’t a simple task. The only tools at our disposal are HTML, CSS and a bit of imagination. Email development forums are awash with questions and suggestions on this topic, plus a graveyard of now-defunct solutions. There’s much trial & error, and the successful method usually amounts to some kind of hacky trick.
Here’s an example. Some versions of the Outlook mobile app will recognise and auto-link dates and times to the user’s calendar. This also turns the associated copy blue. One effective solution is to secretly break up the text with an invisible special character called a zero-width non-joiner. Congratulations – you have successfully tricked an application into losing functionality!
Don’t fight functionality
But why would anyone want to do that? The fact that there’s often no easy ‘fix’ for these ‘problems’ says a lot. The problem does not lie within the application’s functionality. It lies within the sender’s design and objectives.
Suppressing a piece of functionality is not in the spirit of accessibility. And to be frank, it’s not the sender’s decision to make. Nobody likes it when a website blocks or forces the opening of links in new tabs. A similar etiquette applies to the world of email.
Design around it
Addresses are another type of content that could be auto-linked and coloured blue. If they’re sitting on a coloured background, that could result in an ugly clash and illegible text. The solution: place them on a white background instead. Cosmetics do not trump usability.
Outlook has helpfully linked that address to the maps application. Should we break that… or change our background colour instead?
Reallocate the effort
I mentioned trial & error earlier. That means editing code, uploading it to an email platform, sending tests, and checking them on real devices and/or previewing services. All of this all takes time. But this is not a task that deserves it.
Imagine what could be created in that time rather than destroyed. Optimum email designs. Improved accessibility. Better content. Don’t squash the ‘bugs’ – give them a better habitat instead.
Google recently caused a ruckus in the world of email marketing. As part of an update to Gmail, support for background images was (accidentally) knocked out. Oops. The result was an industry of marketers in panic.
Email developers scrambled to find a fix. Workarounds were found, and Google ultimately resolved the fault at their end. Crisis over. This incident will soon be forgotten – which is a pity, as there lessons to be learned.
Things change
This isn’t the first time such an event has occurred. Changes to email platforms are fairly regular. Sometimes for the better, sometimes for the worse. I recall at least two times when a major email platform made a change that immediately broke responsive stacking content on mobile devices.
Or how about some ancient history? In 2007, Microsoft made the infamous decision to switch its ubiquitous Outlook application from a web browser-style rendering engine… to one based on Microsoft Word.
These sort of sudden, unexpected developments vary from subtle to industry-changing. But they have a couple of things in common:
They are beyond our control as email marketers.
The more complex the email, the greater the chance of it being affected.
Ours is a diverse but fragile digital environment
One customer is viewing your email in Apple Mail on an iPhone 14 Pro Max in dark mode. Another is looking at it in the Gmail web app in Firefox on Windows 10. Someone else is using a little-known third party Android app on a flip phone. The point – there are countless devices, platforms, versions and personal settings to cater for.
Now add Outlook and its archaic code support to the equation. With all this in mind, it’s clear why HTML emails can only work thanks to an array of coding tricks and extensive ongoing testing. The more complex the design, the more liable it is to break now or in the future.
Overloading the medium
Like all email developers, I’ve been faced with many moments of hair-pulling frustration. Inexplicable gaps in emails, font problems, wrestling with truncation… the list goes on and on. This raises a question – why are we going to all this trouble?
Thinking specifically about the Gmail background troubles, I cannot imagine any email content in which a background image is essential. Nice, sure. Fancy, sure. But essential, no. As a means of conveying useful information to a customer, a regular image and some text will do just fine.
All of this boils down to the fact that email is far more fragile than a website. And that is not a bad thing. The trouble only starts when we try to force email beyond its capabilities.
Simplicity is key
Most email development struggles are of our own creation. Why battle for hours to achieve a particular design when the easier option is to simplify? This isn’t admitting defeat. It’s making the smart choice to design for the medium, rather than trying to shoehorn a pseudo-website into an email.
Neither does it mean making an ugly email. Simple is not a synonym of dull. A simple email can include static images, and a static image can be as eye-catching and complex as you desire. The email that houses them doesn’t need to be convoluted, and will only benefit from simplification.
Complex email design is less accessible
The hidden beauty of accessibility is that it benefits everyone. The design and coding techniques that it involves will often directly improve your overall email, or serve as a reminder to clean it up.
Complex email design is the enemy of that. It increases the chance of colour clashes, screen reader navigation difficulties and inconsistent use of text and images to communicate information. Simplicity in design means that we don’t have to strive to find clunky solutions to these problems – we circumvent them entirely.
Email code is absurd
It’s easy to forget just how ridiculous email code is. HTML data tables are used for structure. Multiple nested elements are used to achieve something that could be done with a single HTML tag on a website. Spacer objects are often required to force items into place. An assortment of tricks and hacks loosely pins everything together.
And yet we repeatedly choose to attempt complex designs in this environment. Surely the logical choice would be to have less of this clunky code, not more?
Email designers are their own worst enemy (or at least the email developer’s)
Mobile phones have some fairly decent photo editing apps. But they’re no replacement for Photoshop on a desktop computer with a mouse or tablet. The mobile apps are suited to quick, simple edits only. Trying to do anything more in-depth is convoluted if not outright tortuous.
Designing emails that look like websites is like trying to perform complex photo editing on a mobile. It’s simply not the right tool for the job.
Breaking from convention takes courage
Almost every brand sends fancy HTML emails. Companies need to adhere to brand guidelines. No-one wants to challenge the status quo.
That could be good news for you. The one who breaks convention reaps the rewards while others struggle on. Be that one!
Through our extensive travels around multiple email platforms we have noticed that whilst they all do the basics really well, sometimes there is just that missing bit of key functionality that you long for, or it just doesn’t quite work how you want. So, you find yourself doing things the hard way and dreaming of a day where all you need to do is push a button and it all magically happens. But how can you fill the gap?
Well, we thought “why not do something about it?” and came up with our own solutions that anybody can take advantage of. Fortunately, most email platforms provide their own API integrations so these solutions, where required, should be adjustable to suit each platform.
Fill the gap scenario 1: list splitting
You may have a list held in your platform and for whatever reason you need to split it into 2 or more chunks. In theory this is possible with A/B testing but not if you need it for some other purpose.
The hard way: download the data, stick it into a database, run some SQL on it to output however many chunks you want, reimport.
The easy way: use a tool which lets you enter a list identifier, what you want to call the new lists and how many splits you want and it will run the downloading, SQL and reimporting for you whilst you eat that biscuit.
Demo
Here’s a demo of how a tool to split files could work (since it’s a demo we’re restricting to a 1 column file, txt or csv only, and we’ll only split out 100 records):