Loyalty Offers Application

Role: Manager of Product Design & UX Development
Company: Epsilon
Timeframe: 2020-2021
Goal: Enhance the usability of offer creation, introduce innovative functionality to improve Forrester usability score, increase existing client satisfaction and help win new business.

Overview

Epsilon is a leader in loyalty services, helping clients build lasting emotional connections with their customers. Loyalty, part of the Epsilon PeopleCloud suite of products, is a tool that allows clients to build robust loyalty programs, manage their customer base, and create engaging offers that are unique to each customer. It is an enterprise-level loyalty solution used by customers including Walgreens, Sally Beauty and Dunkin’.

The offer creation part of Loyalty is geared towards a specific user - the BSA, or Business System Analyst. The BSA is an advanced, tech-savvy user. Unfortunately, for users other than a BSA, offer creation can have a steep learning curve. Clients were complaining that simple tasks, like creating a “buy one get one” (BOGO) offer, were difficult to complete and only a handful of people on their team were capable of doing the work.

My team was tasked with creating a new application, called Offers, that would allow distributed marketing teams or self-service marketers to create their own offers without having to use Loyalty. I led the project from beginning to end - the process we used is outlined below:

Our goal was to enhance the usability and introduce innovative functionality to improve Forrester usability score, increase existing client satisfaction and help win new business.

Discover

We began our discovery process by assembling a Client Advisory Board (CAB) which included a handful of Loyalty clients who were willing to share their feedback and guide our product roadmap. The goal of the CAB was to inquire and understand the Chief Marketing Officer (CMO) and Loyalty Marketers tasks, needs, and pain points.

We held two rounds of conversations with each client. The first round was a debriefing session with client services, and the second was a semi-structured interview with BSAs and clients services. We spent Q2 of 2020 assembling clients for the CAB, tracking down the right people to attend the meetings, scheduling meetings, writing scripts for our conversations and conducting the conversations. We received very helpful feedback from our clients - we learned all about how their teams were structured, what their daily tasks were, what their biggest pain points were, new features they’d like to see, etc.

Define

We organized our notes from our conversations, and created a powerpoint that outlined the research we did. We shared the deck with stakeholders so everyone understood the outcome of the meetings and knew the direction we were headed. In order to be successful, we would need to address most of the key takeaways in our new application.

Key takeaways:

1. Loyalty provides the technical ability, but is lacking intuitiveness in a self-serve environment

2. The Loyalty offer creation process has a sharp learning curve and requires knowledge of programming

3. Clients require the ability to test different loyalty offers in order to execute the most optimal rewards strategy

4. Clients need the ability to segment customers into specific audiences with Loyalty, so they can seamlessly find and send out specialized reward programs tailored to each member

5. Clients require better reporting tools that can perform ad-hoc analysis with specific segments, in real-time, in order to uncover insights

Design

We started by creating a high-level user flow for one of our most common users.

We would be using visual styles and design patterns from our component library, so we focused on solutions for the key takeaways from our client conversations. Below are the key takeways and how we solved for each.

1. Loyalty provides the technical ability, but is lacking intuitiveness in a self-serve environment

To make the Offers application more intuitive, we focused on simple language and only used terms that were understood by our users. We made sure that on each step of the create offer flow, the user knew exactly what information they needed to input. We wrote clear and concise error messages so that users knew exactly what to fix on the page.

For example, the APIs we were using created error messages that simply said “Required field.” We wrote new error messages that clearly explained how to fix the error.

2. The Loyalty offer creation process has a sharp learning curve and requires knowledge of programming

The offer creation process in the existing Loyalty tool was cumbersome and very technical. They called it the “offer engine” and it was capable of creating any offer imaginable - if only you could figure out how to use it. See below for a screen they shared with us on how to use the existing offer engine:

The offer was written in this complicated language:

Our challenge was to take this complicated offer engine, figure out what most users needed to accomplish with it, and display it in an easy to use interface. We worked with both clients and the sales team to make a list of the most commonly used offers. See below:

We then prioritized the list and settled on ten offers that would become “templated offers”. Instead of an offer engine where the user could create anything and everything, we would provide templates that could be selected and customized further. Below shows the ten offer templates:

As part of this project, we created a set of custom icons to add visual interest:

3. Clients require the ability to test different loyalty offers in order to execute the most optimal rewards strategy

Clients wanted an easy way to test offers so they could figure out what the best strategy was before sending their offer out. To achieve this, we added a sidebar called “Offer Forecast” which had two sections: Customer Count and Offer Simulation. Customer Count gives the exact number of customers who would receive the offer. Offer Simulation allows the user to enter a specific date range and see a simulation of how that offer would perform based on previous transaction history. It would show how many points would be awarded, the projected number of offers issued, etc.

4. Clients need the ability to segment customers into specific audiences with Loyalty, so they can seamlessly find and send out specialized reward programs tailored to each member

Clients made it clear that segmentation was very important because they wanted to create more specific offers that were tied to only a subset of customers. Segmentation is a feature that we have in other products, so we were able to borrow a design pattern that lives in Blueprint and just figure out the details.

In the example below, the user is setting up an offer that will only be sent to members that are in the member tier Gold.

Clients also made it clear that they wanted to be able to use audiences that had been created elsewhere, like Adobe or a different PeopleCloud product. To solve for this, we added the “Select a Pre-Defined Audience” option.

5. Clients require better reporting tools that can perform ad-hoc analysis with specific segments, in real-time, in order to uncover insights

During our CAB conversations, we found out that reporting was very important to our clients and they weren’t getting what they needed. Clients had been receiving a “canned” report from Epsilon that was customized for them, but they wanted more. They wanted to be able to dig into loyalty segments within the application, instead of something they receive in an email. They wanted proactive insights that guided their decision making. They wanted a dashboard they could pull up at anytime, even on their phone, that was easy to read and was helpful for all users - BSAs, CMOs, etc.

We already knew the information that the clients needed in terms of reporting because they were already getting it in the automated email. So we mocked up what it would look like within the application and added places where they could act on those insights. For example, on the Overview tab of the mockup below there are links to “Get back on track” where the application would suggest ways you can increase your daily transactions, or increase the number of active certificates.

Test & Refine

To validate our designs, we set up another round of conversations with the CAB. We walked through the mockups and shared our solutions to their pain points. We asked questions like, “What does the word offer mean to you?” or “What do you think clicking this Simulate button will do?” so we could get their first impressions, or what they hoped would happen, before showing them our solution.

Clients were very excited about the offer creation process being much easier, they understood the general flow, and liked that they could select pre-defined audiences. They asked some great questions that we hadn’t considered yet, such as:

• How was the Online Message tab different than what they had in place currently?

  • It wasn’t any different, but we had to double check and get back to them

• How could they have a workflow across PeopleCloud products? For example, Jen was currently building emails in PeopleCloud Messaging, but Shawn was going to start building emails in Offers. How could they work together, in separate products, and not overlap?

  • We don’t currently have an answer for this, but now it’s on our radar

• Did they still have to use Engine Reload in Loyalty?

  • No they don’t, but we had to double check with a tech lead and get back to them

• Can you duplicate an Offer?

  • We added this feature to the Offers landing page:

Sharing our mockups with the CAB not only gave us the feedback we needed, but it made the clients feel like they were part of the process. They were able to see their ideas come to life and understood that they were guiding our product roadmap.

When I left Epsilon in April 2021, the Offers application was in development. My team of designers and developers were working very closely with product & engineering. UX Developers attended the engineering team daily stand up to make sure features are being built using Blueprint and according to best practices, and designers were attending planning meetings to make sure the team had everything they needed and there were no outstanding questions or edge cases that needed solving.

Here’s a quick click-through of the mockup: