About this project

A "good" click-through rate is 2-5%... It is understandable that people ignore spam, but as a business it is critical to engage with your audience via email.  Whether content is related to marketing, or simply for informational purposes. This project seeks to transform unwanted email into email that people look forward to receiving. 

My role: Research & Design

Research and discovery

My first product design project at Welocalize was to craft and send our product release notes.  The audience included everyone in the company, our partners, and our customers. At first I found this task daunting.  I have to create and send an email to over 10,000 people, including the CEO!  Little did I know that this responsibility would become something that I looked forward to during every release cycle.

Where to start?  I began by looking at what had been done previously.  The prior email was simply a bullet list of features and bug fixes that had been exported from JIRA.  The information was there, but was not in a format that was easily consumable.

I quickly learned that a particular release may only be of interest to a subset of the audience. I also felt that we could improve on how we presented each and every new feature, regardless of the intended audience.

In addition to manually sent product updates, our product automatically sent email notifications to account owners based on activity that occurred within their account. Event-based triggers defined on the backend sent system-generated email to people associated with that event. Here are a few examples:

• Customer had a project that was complete, and their translated files were ready for download. An email notification would be sent to notify the customer.
• Customer had submitted a translation project request.  An email notification would be sent to the appropriate project manager to initiate the work.
• Project manager had submitted a new translation task for linguist to claim and complete.  An email notification would be sent to all linguists capable of fulfilling the request.

Using Mailchimp and Mandrill, I analyzed open rates for both manual and automated email sends.  I both cases I noticed the open rates were approximately 2%, and the click rates were under 1%, which is on the low side.  

Design

I began experimenting with subject line, messaging, and links. I hypothesized that these were the key areas where where an improvement could be made. In the case of product release emails, I added graphics and copywriting that would entice people to engage with the email. I included links to the relevant knowledge articles for the feature. For system generated emails, I interviewed linguists to see what task-related information they cared about the most. I found that the task type, source, target language, scope of work (e.g. word count), and rate were all very important factors. After designing the template and validating it with the linguists, I collaborated with the engineering team to make this data available in the templates via an API. 

Execution (product release newsletter)

Below are before and after mockups of the product release newsletter. 

Retrospective

We began the redesign process by thoroughly analyzing user data. This included tracking open and click rates, user engagement patterns, and feedback on our existing emails. This data-driven approach allowed us to pinpoint areas for improvement and identify content that resonated with our users.

We revamped our content strategy by focusing on user benefits and value propositions in our emails. We crafted concise and compelling subject lines and body copy that highlighted the most important information, making it easier for users to grasp the value of our product updates quickly.

By making these adjustments, open rates doubled to approximately 6%. Click rate rose to approximately 3% which is a significant improvement from 0%. This lead to greater engagement with our product as well as increased efficiency for our linguistic partners.