Editor’s Note: This is a guest article by Chris Gordon, email marketing manager at Noticed. Noticed is a platinum master partner.
The 2019 Ford Mustang has five submodels: Bullitt, Ecoboost, GT, GT350, and GT350R.
“What does this have to do with email marketing?” you ask. Hang tight, we’ll get there.
Each one of those submodels has its own slightly different body build and certain parts only fit certain submodel types.
“But what does this have to do with email marketing?” you ask again, this time impatiently tapping your foot. OK, let’s do the damn thing.
I worked for an aftermarket Mustang modification ecommerce store and to make the email marketing program successful, we had to create bespoke emails for every submodel of every generation of Mustang.
That means for each campaign we sent, we needed to create up to 20-something different versions of it so that a 2002 V6 Mustang owner did not see completely useless 2012 Boss 302 Mustang parts they couldn’t use on their vehicle.
So how did we possibly create 20-odd versions of each campaign we sent twice a week?
Was it by working hard? Sure, but more importantly it was working smart.
Scalability is a popular buzzword in today’s marketing landscape, but it’s also an extremely critical concept to grasp and utilize on an everyday basis.
I learned that very fast in my time in the aftermarket Mustang mod space, and I carry those lessons with me today in the agency world. There, scalability is a really big deal because each client we work with has a monthly allotted budget of hours. So the question becomes, “How do we provide the best possible results in the limited time we have?”
Let me share some of the best ways I’ve learned to make each moment you spend in Klaviyo as efficient and impactful as possible.
Dynamic product recommendations
Dynamic product recommendations were the secret sauce that made those approximately 592,239,392 unique campaign variations at AmericanMuscle even close to possible to execute on.
We started out by manually coding unique product grids for each of the submodels and let me tell you… nobody has time for that noise. But when we started using dynamic product recommendations, it turned that insurmountable task into a “no big deal, put some burgers on the grill, I’ll be home by 5” kind of deal.
In Klaviyo you can get pretty in-depth with how you use dynamic product recommendations. You can use the popular algorithm, which will show your top sellers from a rolling 90-day window. Or you can use the trending algorithm, which is similar but biases the results for recency, giving you a greater variety of products being shown if your store has pretty consistent top-selling items.
You can also turn on personalized recommendations, which will exclude products the customer has already purchased while also tailoring the product selection to their personal tastes by showing them the products they’re most likely to buy based on their past purchase history.
If you’ve ever gotten an email from Amazon, you know how tempting these kinds of algorithms can be (why thank you Amazon, yes, I will purchase that gaudy flame button-up short-sleeved shirt, it will go perfectly with the Guy Fieri Halloween wig I just purchased). Well, Klaviyo makes it possible for you to use the same concept in your emails!
Lastly—and this is important—you can build feeds based on product collections you’ve built on your store’s platform. This means you can do simple things like exclude accessories or build a feed that only shows new products.
But you can go further!
Imagine this: a customer abandons a dress in her cart. You, an intelligent and extremely creative marketing professional, have set up a special abandoned cart flow specifically for customers who abandoned dresses. You lead with the abandoned item and maybe sprinkle in some nice lifestyle shots of people in your dresses, and then below that an automated product feed showing other dresses that may interest your customer.
Maybe there’s a reason she didn’t finish purchasing that original dress: maybe it was extremely purple, like a little too purple. But this dress that appeared in the recommended dress feed is the perfect amount of purple. She buys it and you’ve recovered potentially lost revenue.
Scalable email templates
Once you have your product grid set, what about all that other stuff in the email? How do you scale content?
Templatizing your email structure so you’re not reinventing the wheel with every campaign you send is extremely important, not just for scalability’s sake, but also for results. If every email is wildly different from one another, how do you know what’s working and further iterate on successes?
One of the best ways to achieve this efficiency in Klaviyo is to use saved blocks.
What do you want your CTAs to look like? What about your header? Footer? Product grid? Once you figure this out, hover over the content block and click the “Star” icon on the left. This will allow you to use your creation as a saved block in the future and it will be available with a single click and drag.
Once you’ve created and saved your reusable content blocks, it’s time to demystify the structure of your emails. Building emails with a familiar structure in mind not only helps you be efficient but also helps your customer.
Think about it… you have someone’s attention for only a few seconds every time they open your emails. If they can’t find a CTA or have to search around for a category link, they’ll jump to a faster way to get there and I’ll give you a hint… it starts with “A” and ends with “mazon.”
Here’s an email structure that I’ve found works in almost every vertical. There are obviously exceptions and special email types but generally, for campaigns, this structure won’t lead you wrong (presented in order from top to bottom):
- Header – This should contain the logo (link to homepage) and the most important category links
For our client, Layla Sleep, we developed a simple, effective header that takes up a minimal amount of vertical space so more of the email can be visible without scrolling.
- ASpot – Some people call this a “hero” or “banner” image as well, but it’s basically your main thesis for the email and should link to a shoppable landing page and feature clear headline text and a CTA button (your subject line and preview text should support whatever is featured in the ASpot as well)
This ASpot we designed for our client, Southern Scholar, is effective in showcasing a lifestyle photo and driving urgency in the copy along with a simple, catchy CTA.
- Product grid – This grid could feature products related to the callout in your ASpot or use dynamic product recommendations (just make sure to provide context via section header text, (i.e. “Recommended For You” or “Best-Selling Long-Sleeve Tops”).
This product grid for our client, Smith Farms, features a nice mix of recommended products from a variety of product categories.
- BSpot – This is an additional graphical callout unrelated to what’s featured in the ASpot. This helps to widen the focus of the email and drive more overall traffic.
This graphic comes from an email chiefly about Apparel Candy’s hat category. But, just in case none of those hats caught the customer’s eye, we also provided a graphical callout for their sunglasses category to add variety and an additional shopping opportunity.
- Footer – This reiterates important category links and any other category/evergreen landing page links that didn’t fit in your header.
Our client, HAT CLUB, sells officially licensed professional sports hats for a variety of leagues. The footer is a great place for us to feature the “Shop By League” links that are critical to the shopping experience but would not fit into a header.
Letting the data do the work via flows
Segmentation. It’s important! We can all agree on that. The more personalized we can make our emails, the better off we are.
In some cases, like the case I described before with the Mustang mod fitment, it might be worth the massive amount of extra work it takes to build different segmented versions of each campaign.
However, most of the time, the effort and time it would take to reach this level of handmade personalization simply isn’t worth the investment and you can spend that time more effectively elsewhere.
So you might be wondering why I’m throwing a big bucket of cold water on that email personalization enthusiasm you had… Well, I’m not, I’m just saying manually building a million different versions of each campaign probably isn’t the best way to do it. There are better ways to hit that email personalization nail directly on its revenue-driving head. Let’s go over some!
That highly-segmented campaign you want to build and send? Consider for a second that maybe that segmented campaign might be better off as a flow.
Earlier on I talked a category-personalized abandoned cart email you sent to someone who left a dress in her cart. That kind of email in many cases is built and sent as a batch and blast campaign.
“Let’s send a dress-based category email to everyone who has purchased a dress in the last 60 days!”
Sounds like a great idea, but what if, instead of sending it just that one time, you could send it to dress buyers every single day from now through the end of linear time?
This gets you more bang for your effort clearly, but it also allows you to be even more timely with your send because you get to choose exactly when each individual contact receives the email in their journey rather than having to send to a pooled segment of users at a single time.
The data knows you better than you know you
Let’s talk about another personalization hot topic: user preferences.
Our client, Hat Club, sells officially licensed professional sports team hats. Knowing what sports and teams each customer is a fan of is obviously of critical importance for their success so, for them, collecting that information during the welcome phase of the customer’s journey is a slam run home dunk (that’s a sports thing, right?).
For the Mustang mods company I spoke about before, getting the model and year of the customer’s vehicle was indispensable. And for clothes companies that sell both men’s and women’s apparel, finding out what gender the customer identifies as is also super important.
But besides these types of instances where preference information fundamentally changes the relevant products and categories, most of the time you’re better off determining what the customer is into yourself rather than asking them.
For example, a jewelry company asking if a customer is interested in necklaces, bracelets, or earrings probably doesn’t gather useful information for the long term. Instead, identifying intent and acting on it in a timely manner is a much better strategy.
Did the customer view multiple bracelets within the last few days? I think you might have a bracelet shopper on your hands! A bracelet specific browse abandoned email will likely close this sale by showing them products they showed interest in, plus a few similar bracelets they may not have seen yet.
People are really bad at knowing what it is they want. In fact, most of the time people have no idea what they want until they see it so it’s our job to show it to them. Not only does trusting the browsing data save you the time of building forms and manually organizing the data, but it’s also likely way more actionable and informed.
In most cases, don’t trust the customer to tell you who they are, let them show you.
A word of warning
We’ve had a lot of fun here today talking about all the different ways efficient and scalable forms of personalization can improve your email program, but it’s important to know that personalization also has an evil twin brother named “over-personalization” (their parents weren’t very creative with names).
Let me tell you about how “over-personalization” can work its dark magic to negative effect on your email program with a story.
How many times have you gone grocery shopping while hungry and walked out with seven future-tummy-aches worth of snacks? If the grocery store had asked you specifically what you wanted before you entered the store and only showed you those items, they would have missed out on all of that sweet snack revenue.
My point is: personalization should never impede discovery.
Overuse of segmentation and personalization strategies leads to a very insular experience for the customer. They will only see the things you think they’re interested in and they will never get a chance to discover new products or categories they don’t yet know they want.
About a year ago, I bought a vacuum from Amazon and Amazon’s personalization algorithm apparently went “THIS GUY FREAKIN’ LOVES VACUUMS” and showed me nothing but vacuum recommendations for months.
Obviously, this was not a great experience for me and who knows how many potential midnight impulse purchases they missed out on by showing me a million vacuums instead of allowing me to discover new things I might like.
Respect your own time. Not only for your own sanity and happiness, but also because every second spent painstakingly hand-building campaigns or manually managing hundreds of customers segments is time not spent on potentially more impactful strategies for your business.
If something is going to take you a long time to accomplish, ask yourself if it will meaningfully move the needle for your business. If the answer is no, find a faster or more efficient way to do it.
I hope the tips I went through in this article will help you to do exactly that. And if you’re stumped, give us a call at Noticed. We’re clearly big email marketing nerds who think about this stuff way more than any reasonable human should, so let our pain be your gain. We’d love to hear from you!
Curious how your email marketing stacks up against your peers? Check out these ecommerce email marketing industry performance benchmarks.
Back to Blog Home