Creating a subscribe and earn program — from scratch

Context

It started as a pitch

PC Express is the grocery e-commerce platform for Loblaw, serving over 3 million active customers and generating nearly $2B in annual revenue. Alongside it sits PC Optimum, the largest loyalty program in Canada.

Subscribe & Earn started as an idea during a pitch competition to our VPs—it wasn't my idea, but I participated and had other proposals (please enjoy the picture of my team below). The idea set out to reward customers with PC Optimum points when they subscribed to items and received them every week, two weeks, or month. The hypothesis was that subscriptions could increase retention and repeat purchase behavior by removing friction from habitual shopping.

Research

Exploring existing subscription models

I led the design for the program on the UI side, and worked closely with Operations teams to understand impacts to existing systems and design new processes. But aside from having a fantastic team with me, we had almost nothing to go off of; existing research was limited, time was far too limited. I worked with my UXR team to get some insights collected as I ran a large-scale comp analysis into subscriptions.

From researching subscription programs, they fell into two categories: subscribe now (akin to Amazon), or an add to cart to subscribe model (like Safeway). We found both approaches interesting and possible options for PC Express.

Design

Testing two distinct models

I went deeper. I mapped out what each model could look like from a flow perspective, keeping in mind existing tech and business constraints that I knew of. My PM, business partners, and myself all assumed the subscribe now model was cleaner, but it had too many dependencies on our existing business rules with PC Express; in addition, this model meant we'd have to recreate the checkout flow on each item's PDP, rather than consolidate with our existing checkout flow.

I mocked up both and tested with customers to see if there was a high variance in responses depending on tasks, but ultimately there weren't.

This insight helped us land on a cleaner design, with fewer tech constraints and rework, and it allowed me to continue mapping blueprints with my Ops teams.

Challenges

Business constraints impacted our strategy

Unfortunately, late into the project, we learned about business rules that would almost completely defeat the purpose of the program. To avoid high fulfillment costs, all orders would require a $35 minimum order value, including subscriptions. In addition, multiple orders couldn't happen at once, no matter the size.

I presented the challenge to my PM and business partners, and explained what this would mean for the program. I then spent the week ideating on possible solutions — the first was merging subscriptions with other open orders, and the second was consolidating subscriptions so they all happened on the same day of the week, to maximize overlap of subscriptions and thus order value.

Outcomes

Launch and results

The launch was huge. The marketing push was intense, and the marketing team offered high PC Optimum points rewards at launch to incentivize sign ups. As a team, we were able to provide value to Canadian customers through this solution, despite navigating tough tech and business constraints.

* Please reach out to me for details on numbers.

Hypotheses and learnings

I hypothesize that we could’ve learned faster and got most of the metrics we were looking for through a slimmer solution. If I were to do it again, I think we should've started with an auto-add to cart feature, and used that data to inform how many people would be interested in actual subscriptions.

Aside from success, my way of working also changed as a result with this project. I went from having a lot of meetings, to just recording videos async to get feedback faster. It helped the engineers, PMs loved it, and it helped us create a traceable thread for decisions that we could refer back to.