Bundle Upsell
Hooked on Phonics is a monthly subscription reading program for children between 3–8 years old. For $16.98 per month, users get digital app access and are shipped companion physical materials. Retention rates are not terrible, but not stellar either.
The Objective
Increase LTV (customer lifetime value) while also providing value to the customer.
The Challenge
Drive incremental revenue by offering the entire program in the primary acquisition funnel upfront to those who have less price sensitivity while not adversely affecting conversions of the core monthly subscription offer.
Previous State
In the primary acquisition channel, only subscription plans are available.
The top Sellers on the Shop page are larger upfront bundles.
Customer Service fields questions regarding purchasing the entire program all at once. It’s not a huge number, but not significant either.
Known Knowns
Bundles are Top Sellers in Shop
The Shop page offers the entire breadth of product offerings and configurations. Bundles are the top-selling SKUs.
Churn & Price Sensitivity
The tops reasons for subscription cancellations revolve around price and value.
What’s Included?
Per user research, most users don’t know what exactly is included in the program. Most are compelled by the $1 trial offer for the subscription.
Considerations
Subscription CVR
Could proffering a discounted bundle adversely affect conversions as a whole? Will it come across as too aggressive?
Fulfillment & Bundle Kitting
Do we have the various bundle configurations pre-kitted or do they need to be picked and packed?
Payback Window Shortened
Incremental bundle revenue has a drastically shorter payback window than subscriptions and with no possibility of churn.
Testing Inventory
How many variants can we test to reach statistical significance in a reasonable amount of time?
Recurring Price Transparency
Could showing the recurring subscription price, which we don’t currently do, adversely affect subscription conversions?
Upsell Suppression
Incremental bundle revenue has a drastically shorter payback window than subscriptions and with no possibility of churn.
Bundle Sizes & Prices Vary
Program bundles are configured from where the child is placed. As a result, bundle sizes, prices and discounts vary. Could that affect take rates by grade?
CURRENT STATE USER TESTING
“I ‘m not really sure what’s included, but $16.98 per month sounds reasonable. I thought it would be more. That’s my coffee budget for a week.”
— Anna, user testing participant
Success Metrics
Total Revenue
With a new revenue model added to the mix, Total Revenue, in aggregate, would be tracked over the course of the average retention period.
Lifetime Value (LTV)
LTV over the course of the average retention period would be tracked in aggregate and separately by Subscription vs. Bundle takers.
Bundle Take Rate
Aimed for a 5% of all orders to be Bundles. This is in aggregate across all grades levels.
Conversion Rate (CVR)
Retain the CVR for Subscription takers or take a slight drop in CVR and make up for it with Bundle revenue.
Churn & Retention Rates
Increase our retention rate by 1 month, thereby reducing our churn rate as well.
The Process
-
After a concept is generated, key questions and known unknowns are formulated. Existing data is identified and scoured to help answer those key questions. These may come in the form of:
Traffic logs
User surveys
Customer service requests
Previous user research
Previous and relevant A/B tests
Marketing analytics
From there, any gaps and relevant research methods are identified before crafting a research plan.
-
Informed by A/B testing results, clickstream analysis, session recordings, and moderated user testing, new UI concepts are created to help mitigate any monetary and UX performance shortcomings.
From there, new concepts are user tested before moving anything into development.
User research methods used:
Qualitative attitudinal methods: User interviews
Qualitative behavioral methods: Moderated user testing, session recordings, heat maps
Quantitative attitudinal methods: User surveys
Quantitative behavioral methods: A/B testing, clickstream analysis
-
Experiment designs are crafted in close collaboration with product management and marketing.
Testing plans include:
Success metrics
Number of variants
Testing inventory needed
Traffic allocation
Experiments are then slotted into the experimentation roadmap.
Test Results
Round 1 of the test is ongoing (as of July 2025). Here is what we see so far:
Revenue estimates are negative 👎🏽
Subscription CVR is down significantly 👎🏽
Average of 7+ clicks per user 🖱️
+5.63 more interaction events🖱️
What’s Next
-
Moderated User Testing
Add color to existing quantitative & qualitative data
-
Less Intrusive Bundle Option
Retain CVR for subscriptions while still offering a bundle
-
Another Round of Price Testing
Home in on the winner by reducing the price ranges
-
Different Bundle Configurations
Offer 1 or 2-year bundles instead of the entire program
