Introduction

Project Overview
A delivery promise is a statement shared with customers related to how and when they should expect their order to be delivered. Our Delivery Experience team set out in 2023 to discover how we might design and deliver a better delivery promise across multiple teams roadmaps in 2024. I conducted discovery research to inform the development of our business requirements and created concepts to align design and product decisions across several product teams.

Role
Staff Designer for Delivery Experience:

Discovery research
Concept development
Alignment with other product managers and designers across the core product team.

Team

Content Designer for Delivery Experience
Associate Director Product Management for Delivery Experience
Designers and Product Managers for each of the following teams:
- Product Listing Page
- Product Description Page
- Care and Checkout
- Recommendations
- Autoship (Subscription Program)
- Technical Product Manager for Promise Service API Team

Problem Statement

The Challenge
Today, competitor online retailers have shifted customer’s expectations to see explicitly when they should expect to receive their delivery. Chewy’s current site only communicated generic language about delivery times that did not take into account the customer’s location or operational data related to the supply and location of products on offer. The Delivery Experience team was challenged with understanding how this change in communication would affect Chewy’s customers in order to define requirements for building this new feature and gaining alignment across multiple teams to deliver on the following year’s roadmap.

Competitive Analysis showcasing other retailers solutions for Delivery Promise

Context and Constraints
Prior to the Delivery Promise initiative, Chewy did not offer customers a pre-purchase delivery promise. Customers were not meaningfully notified of when to expect their package until they received a shipment confirmation email. A team developing the calculation of delivery promise for operational purposes committed to delivering a delivery promise calculation by Q1 of 2024. However, in order to ensure that they would be able to deliver something that would be meaningful for consumers, they needed complete customer experience requirements to guide and prioritize their work. We conducted a discovery to help define the initial concept for delivery promise so that we could seek alignment with the multiple teams during Chewy’s yearly prioritization process, called OP1 (standing for Operational Plan). Chewy teams work through this process to decide all the work that will be done in the following year. If dependent teams do not agree to complete the work, the project will not be prioritized and will have to wait until the following year.

Research

Alignment
To kick off our discovery I held a learning agenda workshop with our stakeholders to discuss their concerns about the delivery promise initiative. Together we documented possible risks to the projects, and questions to answer about how to effectively deliver on the project. We then prioritized each of these items into a formalized agenda to inform the research to follow. I then categorized each of the items according to the most appropriate research method to answer the question, and applied a status tracker to each of them to ensure we were on track.

Confluence page for tracking our learning agenda goals.

Research Goals
The primary aim was to explore how customers perceive and interpret the delivery promise, which includes accuracy and expectations regarding delivery dates. This research would help guide the upcoming experiments and inform future improvements to the customer shopping experience.

Methodology

Participants
Fifteen participants were recruited, divided into three segments: Chewy customers, Non-Chewy Pet Parents who shop online, and Churned customers (those who haven't purchased from Chewy in over a year).

Interviews
Each participant took part in a 30-minute structured interview conducted via Usertesting.com

Discussion Guide
Participants were asked about their online shopping habits, specifically focusing on their expectations and experiences with delivery promises. They interacted with the Chewy website to identify delivery-related language and compared it with competitors.

Findings

The research revealed significant insights into customer perceptions and behaviors regarding delivery promises. The findings indicated a need for clearer, more accurate delivery information to help customers plan better and trust the retailer’s commitments. These insights will guide the Delivery Experience team in refining Chewy's delivery promise strategy and enhancing the overall shopping experience for customers.

Key Insights

Visibility and Interpretation of Delivery Promise
- All participants easily found the delivery promise "FREE 1-3 day shipping over $49" on Chewy's product pages.
- However, the wording was ambiguous, leading to varied interpretations about what happens if the shipping threshold is not met.
Ambiguity and Clarity Issues
- Participants generally understood the concept of a free shipping threshold but were confused about the specifics of delivery versus shipping
- Many were unsure about the exact delivery time frame and whether weekends or other factors would affect the delivery speed.
Importance of Delivery Date Accuracy
- Most participants prioritized an accurate delivery date estimate over speed. Knowing exactly when an order would arrive helped them plan, especially in urban or apartment living situations where receiving deliveries can be more complex.
- The Churned segment also valued accurate delivery dates but had a slight preference for speed when immediate needs arose.
Perceptions of Free vs. Paid Shipping
- Participants were willing to wait longer for free shipping and were more patient with delays. Conversely, they had higher expectations for paid expedited shipping and were likely to seek refunds if these expectations were not met.
- Free shipping was often used strategically to avoid additional costs, with participants meeting thresholds to qualify.
Factors Influencing Delivery Expectations
- Prior experiences with retailers played a significant role in shaping delivery expectations.
- Participants expected faster deliveries from well-known retailers like Amazon and adjusted their expectations based on their past experiences.
- The type of product (e.g., heavy items, luxury products) influenced perceived delivery speed, although this did not significantly affect expectations for Chewy products.
Consumer Behavior with Late Deliveries
- Late deliveries were common but generally accepted as part of online shopping. Responsibility was often attributed to third-party carriers rather than the retailer.
- Proactive communication from retailers about delays was appreciated, and lack of it led to assumptions of package theft and increased customer service contacts.
Participant Preferences for Delivery Information
- Product description pages were identified as the most critical place for displaying delivery promises.
- Participants favored a single, accurate delivery date over a range, preferring conservative estimates that met or exceeded expectations.
Competitors
- Amazon was consistently mentioned as setting the standard for delivery expectations.
- Participants often viewed delivery dates from these retailers with skepticism, interpreting single dates as the latest possible arrival time.

Concept

Scope

Following our research readout, I set to work on identifying the key areas across the customer purchase journey that would be affected by a pre-purchase delivery promise. They were as follows:

Product Listing Page (a.k.a. Search Results)
Product Description Page
Recommendations
Cart
Checkout
Thank you page
Order Confirmation Email

Refining the Message

In conversations with our tech team, there was some question about their ability to deliver an explicit date for the MVP version of Delivery Promise, so they asked us to provide some additional messaging options for consideration. I partnered with our content designer to arrive at a prioritized list of how we would ideally display dates to the customer, with the first being our ideal state:

Explicit Date “Arrives August 25”
Explicit Date Range “Arrives between August 22 and August 30”
Number of Days “Usually deliver in 5 days”
Day Range “Usually delivered in 4–7 days”
Non-dynamically generated day range (current state): “Delivered in 1–3 Days”

This gave the tech team some considerations about what their API would deliver to front end teams, and allowed them to discuss how the underlying machine learning model’s performance would affect the customer experience, such as the tradeoffs between accuracy and precision.

Concept wireframe showing Delivery Promise on the product description page.

Capturing Customer Location

Our Delivery Promise model required three key inputs to make a calculation and return an output, the product SKU (stock keeping unit) number, the selected quantity of the product, and the customer’s location. The first two of these were easily made available on the pages of the Chewy site, but the third was a bit more complicated. While it would be conceivable to make assumptions about a customer’s location according to their default shipping address, it would only be possible to use this information for logged in, or authenticated customers. This left out a significant number of unauthenticated users, for which we would need some method to capture their location prior to providing a delivery promise.

Concept wireframe showing 1. Modal Interaction and 2. Alphanumeric input for US and CA

Providing an affordance for these users would also have a benefit for authenticated users, allowing them to flexibly change their intended shipping location without having to add a completely new address into their account settings.

Prior to receiving input from the customer, we needed to provide a default call-to-action to let customers know the value of providing their location. Once a location has been provided, then the customer would be able to see the Delivery Promise.

Layout

Based on my research, I knew that the current location of Chewy’s delivery statement near the “Add to Cart” call to action button was currently working well for customers, so the ideal approach to incorporating a delivery promise was to use that same space by simply editing that message. We found customers conflated two distinct ideas in the “Free 1-3 Day Delivery over $49” message. The first idea is that delivery is free over $49 is what we call a shipping threshold. The second idea is that delivery will take between one and three days.

Chewy’s current “Buy Box” showing the large orange “Add to Cart” button and delivery statement.

The shipping threshold is important because customers are very hesitant to pay for shipping if they don’t have to, and many opt to simply shop in bulk to avoid paying for something that doesn’t offer any additional value. In order to alleviate this confusion, my first recommendation was to split the delivery promise messaging from the shipping threshold message, but maintain their proximity to one another.

Recommendations

Some internal stakeholders had voiced concerns following our research readout about displaying a promise that was longer than our current 1-3 delivery statement. According to our internal metrics, approximately 10% of Chewy packages took longer than that time to be delivered. These late deliveries have a contact rate of 2.85%, which is 6X higher than normal. We had to find a solution to address concerns that customers would likely abandon their carts if they found an item was going to take longer than their expectations, and so we settled on incorporating recommendations as part of the delivery promise work as well. If an item was going to take longer than typical, we would surface a recommendation, by placing an anchor link directly beneath the delivery promise.

Handoff

Following concept development, I held several sessions with the design teams responsible for delivering to discuss the discovery research and conceptual mockups to ensure they had the appropriate context and direction to take the work forward, and as of this writing, the teams responsible for the product description page, as well as the order confirmation have experiments in flight, with results forthcoming.

Delivery Promise