Skip to content

02/Case study

Same-Day Delivery

Telling customers in four newly launched UK cities that Same-Day Delivery had arrived, and learning where it worked.

Company
Amazon Prime
Year
2025
Role
Owner: campaign, experiment & EU playbook
Category
Product adoption · GTM

01Context

Same-Day Delivery was launching in four new UK cities: Aberdeen, Darlington, Plymouth and Middlesbrough. A faster promise only matters if the people who can use it know it's there.

02Opportunity

Customers in those cities had no reason to expect anything had changed. The question was whether a local, well-timed message could get them to try Same-Day, and whether it would work the same everywhere.

03Approach

I set up a city-specific app push campaign. Each city got its own headline ('Plymouth, want it today?', 'Aberdeen, get items today!') linking straight to the Same-Day storefront.

To avoid promising something we couldn't deliver, I only targeted postcodes with full Same-Day coverage. Each city ran as its own 50/50 test-versus-control experiment, so we could compare results city by city.

04Execution

I handled the targeting, the creative, the QA of every link and storefront per postcode, and the launch. Then I measured the impact during the campaign and over the following two weeks.

From the results I wrote the playbook for future launches: when to trigger a campaign (speed improvements, new categories, seasonal events), how to target it, and how to set it up step by step. It fed into the H2 2025 roadmap and was written so other EU markets could reuse it.

Selected creatives

Push notification: Plymouth, want it today? New: Same-Day delivery is now available in your area
City-specific push, Plymouth

05Impact

Mobile app purchasesMiddlesbrough · 2 weeks
+24.6%
Push tap-through rateBenchmark 5%
5.2%
Opt-outs from push
0
UK cities, each with its own test
4

People engaged with the messages: tap-through came in above benchmark and nobody opted out of push. What happened after the tap depended on the city. Middlesbrough saw app purchases rise 24.6% over two weeks, and Aberdeen saw a 1.2% lift in Same-Day buyers. Plymouth and Darlington were flat to negative.

06What I learned

A single average would have hidden the most useful finding: the same message works differently depending on the city. The playbook now starts by grouping cities by type (size and shopping behaviour) before choosing the channel and message.

Figures are limited to relative results I'm able to share. Internal volumes, spend and customer data have been left out. Creatives shown were customer-facing.

Next project

Prime Exclusive Deals

Amazon Prime · 2025