Redesigning Re-subscription Strategy

Marley Spoon sends weekly meal kits, customer select recipes, and the company delivers fresh, seasonal ingredients to cook at home. Marley Spoon had around 300,000 active subscribers worldwide. During my time there, I led the design of a subscription reactivation flow for Marley Spoon and its sibling brand Dinnerly, improving conversion from 4.5% to 5.6% in key markets (the U.S. and Australia) between 2022 and 2024.

Problem

Users who deactivated for more than three weeks landed on an empty page with no upcoming orders shown.

Many meal kit customers switch brands often, only 10% stay loyal to Marley Spoon. After three weeks of inactivity, we disable recipe generation, leaving returning users confused and uninspired to reactivate.

User Research

I interviewed 11 former subscribers who has unsubscribed in the past six months.

Each session lasted about 45 minutes and included hands-on reactivation tasks. I asked questions about their reasons for leaving. Then participants navigated the reactivation flow and reviewed Marley Spoon' reactivation marketing email.

Insights

Most users found the reactivation steps easy to understand, but few wanted to return.

Many anticipated using a discount temporarily and then canceling again. Some users declined to return due to unclear pricing, missing details, or unappealing menu options.

100% of the participants subscribed to multiple meal kit brans at the same time. They indicated that upon receiving a more attractive offer, they would unsubscribe from one brand and switch to another.

55% of the participants deactivated their subscription due to high costs. The rest of them said that they stopped using the service due to unappealing options, life changes, poor ingredients, and delivery issue.

36% of the participants would return and leave after using the discount. The rest mentioned they wouldn't use the service again unless their issues were addressed.

45% of the participants were certain that by reactivating their subscription, they would have to pay for their upcoming delivery immediately. The rest didn't know when they would pay for their upcoming box.

90% of the participants appreciated the promotion but found it confusing. They were unsure if it was a new discount, particularly if they had already used one when signing up for Marley Spoon.

64% of the participants expressed an interest in new customizable and discounted recipes. The new offerings sparked their curiosity and could be a factor in their decision to return and use Marley Spoon again.

36% of the participants would not reactivate their subscription due to the absence in pricing details. The lack of information on discounts and taxes can prevent them from reactivating their account.

Stakeholder Workshop

I brought product, CRM, marketing, data, and customer service together to identify opportunities.

We mapped pain points and ideas from each team's perspective. The workshop helped us agree on which features would solve the most urgent problems and deliver measurable improvements.

Conceptualization

I designed a landing page that showed weekly menus, pricing, and subscription details before reactivation.

85% of returning users came back after more than four weeks. By that time, most had forgotten their previous subscription details. The new preview reminded them what they could expect and gave them enough context to make a confident decision.

Prototype

I built a clickable prototype for mobile and desktop to simulate the new flow.

On mobile, I placed a sticky cart at the bottom with price and a clear CTA button. Because 94% of users reactivated through promotional links. the cart always showed the discounted price. Tapping the button revealed a summary of the upcoming order.

Multivariate Test

I tested three versions of the flow to see which design produced the best results.

Variant A was the control. Variant B focused on weekly menus. Variant C emphasized subscription details and pricing. We tracked conversion, engagement, and reactivation-to-delivery rates to compare performance across versions.

Version A
Version B
Version C
Version A
Version B
Version C
Version A
Version B
Version C
Version A
Version B
Version C

Impact & Performance

Displaying the weekly menu did not result in a higher reactivation rate.

28% of users exposed to variant B, featuring weekly menus, experienced lower reactivation and reactivation-to-delivery rates. Our assumption was that although users may initially be excited by the menu, upon viewing it, they may choose not to reactivate their subscription if the week's recipes are not suitable for them. Ultimately, we chose variant C due to its superior overall performance.

+4.5%

In conversion rate* in the US over the period of 3 months

+2.7%

In reactivation-to-delivery rate** in the US over the period of 3 months

+5.6%

In conversion rate* in Australia over the period of 3 months

+2.1%

In reactivation-to-delivery rate** in Australia over the period of 3 months

*Conversion rate measures the percentage of users who click on a CTA to reactivate their accounts. **Reactivation-to-delivery rate measures the percentage of users who receive a delivery after reactivating their accounts.

Multivariate Test

To ensure we provide the most effective content for subscription reactivation, we've decided on a multivariate test to determine the most impactful design solution.

In the multivariate test, we evaluated three variants of distinct design solutions: the control design (variant A), variant B featuring a weekly menu, and variant C presenting service information and subscription types. Alongside reactivation rate, we tracked engagement metrics to comprehend user behavior when interacting with menu and service information, aiming to understand how this information impacts the reactivation rate.

Impact & Performance

Displaying the weekly menu did not result in a higher reactivation rate.

+4.5%

In conversion rate* in the US over the period of 3 months

+2.7%

In reactivation-to-delivery rate** in the US over the period of 3 months

+5.6%

In conversion rate* in Australia over the period of 3 months

+2.1%

In reactivation-to-delivery rate** in Australia over the period of 3 months

Design Approach

Research

UI/UX Design

Multivariate Testing

Competitor Analysis

I reviewed how competitors handle reactivation flows.

Every competitor I looked at directs users into "Select Meals" with a single click. Meals options were shown to users to attract them during the reactivation process. Users reactivate first, then customize the plan, skip the cart, or change delivery schedule afterward.