Send Window Optimization
Data-driven send timing without tripling the work
- My role
- Product Designer
- Timeline
- March – April 2026
Context
Our current send time optimization flow asked marketers to build the same broadcast three times and left them unsure when messages would send.
Project
Enable marketers to use data-driven send optimization without tripling their creative work, and surface the information they need to send without legal uncertainty.
Because of the friction, marketers were abandoning the feature entirely, leaving campaign improvements on the table.
The process
- 1Empathize
- 2Define
- 3Ideate
- 4Prototype
- 5Test
Empathize
To start, I analyzed customer feedback collected from our AI meeting recordings.
EmpathizeCHALLENGE 1
Triple work.
Setting up one broadcast with send windows was triple the work of a broadcast without it. Our marketers manage multiple channels at once, so tripling their workload on just one channel wasn't worth the tradeoff.
“For one broadcast you do have to create 3 because you're creating an AM, mid-afternoon, and then PM of your group audience.”
“[Send windows] had some good adoption when we were pitching it several years ago. It was just… kinda inefficient.”
“We unfortunately do not have the capacity to be setting up multiple different SMS broadcasts... Will there ever be an instance in the future when send time optimization is just a button we can click?”
EmpathizeCHALLENGE 2
Lack of clarity.
Marketers couldn't see when their broadcasts would actually reach subscribers, which led to anxiety about sending outside legal hours. With text messaging's strict federal compliance rules, this wasn't a minor inconvenience but a genuine legal risk.
EmpathizeCHALLENGE 3
Hard to find.
Send windows were switched on with a filter buried among 30+ others, so we relied on Customer Success to pitch the feature. Marketers who never had that conversation missed out on a feature that could boost their CTR and revenue.
Define
Next, I defined our user and their needs.
DefineUser Persona
Melissa is the face of our user. She's busy and always looking to streamline her processes. She is motivated by improving her results, but needs to have trust that our recommendations will improve her results.
This persona graphic was made with the help of our in-house UX researcher.
DefineEmpathy Map
To understand the current process, I explored the feeling of the user as they go through the flow. I found that the more rework we required, the more frustration and uncertainty.
Ideate
Now that I understood our user, it was time to ideate.
IdeateHow Might We?
- How might we...
eliminate the re-work caused by creating three broadcasts?
- How might we...
use tooltips and labels to explain this feature?
- How might we...
surface what we're getting from the “black box”?
- How might we...
encourage organic adoption without relying on CS?
IdeateDecisions
The key product decisions that shaped the build.
- A single toggle
Shows on or off at a glance, takes one click, and matches other optional features.
- AI purple and sparkle
Signals a data-powered feature, using the same pattern as our other AI tools.
- Right below send time
Surfaces the feature at the exact moment marketers decide when to send.
- Show all three times
Three purple time boxes replace the black box with exact send windows.
Prototype
Now that we had the flow, it was time to start designing. I used existed design system components and created any new necessary components to create high-fidelity mocks and a Figma prototype.
PrototypeHAPPY PATH
PrototypeUNHAPPY PATH
FETCH ERROR
SAME-DAY SEND
PrototypeEDGE CASE
Test
After my first iteration, I tested with potential users.
BEFORE TESTING
AFTER TESTING
Our initial assumption was that users would only want to use our algorithm-recommended time for their segments. After some testing, we found that users wanted the ability to adjust these times after the campaigns were created.
Wrap up
Wrap upResults
↑5%
CTR increase for brands using send windows, 90 days after launch.
↑100%
Increase in accounts using send windows, 90 days after launch.