Email lifecycles are completely automated and they should be data driven with visitor events & business intelligence data. With that in mind, you can “growth hack” your way to better lead generation or revenue with some unorthodox techniques.
You might be hunting for growth with traditional lifecycles such as cart abandonment emails, browse abandonment emails and welcome emails and this might not be enough. In this article, we’ll cover some lifecycles that will give you the following:
- More engagement with your customers
- Emails that will help them find the product they want to buy, even if they aren’t aware of it!
- Emails that will help your program by driving higher open rates, high conversion rates and more sales
Two of such email lifecycles we will cover in this section are Search Abandonment emails and Product Lifecycle emails.
Search abandonment emails
Search abandonment emails come within the group of ‘activity’ abandonment emails and can lie further up or down the funnel in terms of the probability/intent of purchase than its two counterparts, browse abandonment and cart abandonment emails.
Site search result pages is one of the highest converting pages on most online retail sites. Users that are searching specifically for products are in the late stages of buying and consequently convert at a higher rate than most normal visitors do. According to research by Econsultancy, when a search tool is implemented into a retail site visitors converted at 4.63%. This is 1.8 times higher than the normal rate, 2.77%.
Similarly, auto suggest is one of the most liked options by customers. Other research by Econsultancy found out that 25% of the users click on the search suggestions. This is especially true in the case of electronics, where there is a much higher variety of products with serial numbers and names that are difficult for an average user to remember. This is a huge untapped ‘learning insight’ for the retailers that most marketers often overlook.
Now that we know that site searches perform extraordinarily well for us, why do we choose to ignore the group of customers who searched for products but later abandoned their search/browse? Why not go a little out of the way and design a lifecycle that targets such users?
Find out the best 5 or 10 search terms that are performing very well for you. Whenever a user searches for these ‘high-performing products’, identify the user through a cookie and register the email address to a group such as ‘search population’. Suppose the user abandons without adding the product to the cart or without even browsing the product (maybe the user wasn’t able to search for the exact product of his/her own choice), fire an email lifecycle of a single, double or triple email series with the search results and cross-sell products (optional).
Carry out multiple A/B tests with the subject line, the body copy, personalization level of email, the frequency of the email and the response time to see what gives you the best return.
In most cases when users decide to leave after using the search bar, the search terms are very specific products rather than a large group of products. If they are searching for a category or something broad instead, you can also design and build a broader search abandonment email.
- The subject line of search abandonment emails should be the utmost priority. This is because you don’t know for sure if the user is certainly interested in the products and whether or not he/she actually found the items they were searching for.
So try to go with a compelling subject line with something that piques their interest. Also, address customers directly by using words like “You”. You may or may not call them on a first name basis.The below email example is sent by Calvin Klein to a user who searched for shirts/dress shirts but did not proceed to the product page. The email used a subject line saying ‘We thought you may like these’. Nothing elegant or click-bait like – this is just a simple email that recommends products to the user.
- When it comes to the email body, your content must be relevant to whatever the user searched. Calvin Klein started their email with the headline ‘Still looking for Shirts + Dress Shirts?’ The user will be able to relate to his/her need.
Besides, the email must contain actual pictures of products that the user searched rather than a simple CTA leading back to the page.
In the above example, Calvin Klein presented users with 3 options of the category searched. Below that, the user is presented with some recommended options. There are separate CTAs for each of these.
Cubavera, on the other hand, goes a step ahead and displays a large number of products in their search abandonment email. Each of these has an individual CTA that leads them to the product page.
In our own testing, product links consistently are in the top three in terms of clickthrough rates so it’s best to test this out for yourself.
Generally, it is not advisable to incentivize your emails that are early on in the sales funnel. Customers display a varying intent of purchase which may range from nil (they were simply window shopping or checking the prices for products they intend to buy at a store later) to high (the user has a specific product in mind and if presented the right one, will go for a purchase). As such, there are no guaranteed lift you can count on upon providing incentives to users.
Example: a customer who is nearing the expiration date of a certain software and is searching for alternatives to the one used beforehand. If presented with the right software and offer, the user may go for a purchase.
Product Lifecycle emails
This is another method of email marketing our way to higher sales and revenues by exploiting the unique behavior of customers. At a brief glance, customer behavior appears as highly chaotic and deriving a meaningful relationship can be difficult.
But upon proper organization and segmentation of customer data, we can run correlation analyses and predict the users’ next purchase. This helps us to retarget the exact customers with the right offers at the precise time. This process can be interpreted as a spin-off of cross-selling with the difference being that it does not merely use categorical recommendations. Rather, it uses customer behavior which is often very unambiguous and changes frequently.
For an example, consider a group of customers that purchases faucets from you. A small percentage of this group then proceeds to purchase shower curtains. You know now that this group is remodeling their bathroom. You can add such customers into a common group and send them emails with high product affinities such as soap dishes or bathroom lights.
We ran an analysis for CellularOutfitter to predict the users’ next purchase – similar to the one we ran for cross-selling.
The full analysis was covered in part-3 of the email life cycles. We had used the added _from URL to segment the products into two main categories- phone cases and wallets/wrists/clutches. But this division turned out to be pretty broad and depending on how the user navigated to the product page, a given product could fall in either of the two categories.
To refine the results of our analysis we worked with the product team to come up with segments within the product category “Phone cases” which was more granular than the earlier version but was still broad enough for us to make meaningful conclusions from the data.
We compared categories of products ordered by customers in their 1st order and in the immediate next. The analysis revealed that customers who had ordered a certain category of phone case in their 1st order are most likely to order from the same category or from a similar category in their next purchase. See the charts below for details about all the categories and the strongest correlation between consecutive orders.
The strongest correlations are shown in green color.
Next, we tried to derive a relationship between the product mixes of the 2nd order conditional upon the 1st order containing a specific category. As expected, products of the same category have the strongest correlation. But we also came across different categories bearing a strong correlation with a specific category of the 1st order.
The chart below shows the product mix of the 2nd order (shown in rows) subject to the 1st order belonging to a certain category (given in the columns). The first column is the baseline product mix for the 2nd order and the subsequent columns show the 2nd order product mix when the 1st order is a product from the category shown in the column header. The delta column represents the difference versus the baseline mix. The 2nd order mix does not sum to 100%. This is because the calculation is binary- either yes or no depending on whether the 2nd order contained a certain category. So an order with multiple phone cases in different categories will be included in each of those categories.
The strongest correlations are shown in green. The results did show us quite interesting results- a customer who had ordered a flip case in the 1st purchase was most likely to buy a flip case again, but also had a good chance of purchasing a hybrid case.
This comparison of category mix versus baseline is still pretty high level, so we tested for significance for some of the larger deltas (using Chi-Squared test which is best for categorical data sets) and most were statistically significant at 90% confidence.
Based on this analysis, we retargeted existing customers with similar types of phone cases that they purchased in prior orders. We created an email list of existing customers for each category (e.g. wallet cases). We then sent those customers emails for new releases and promos within that specific category (or related categories). We created different campaigns on Facebook to retarget these users through their email addresses.
Have you used any such unorthodox technique to target customers? I will love to hear from you.