Sprout Social

Problem statement

How might we evolve Sprout Social’s compose experience to match the mental model of how sophisticated social media managers plan cross-network content?

My role

While I’ve been purely a manager for most of my 4+ years at Sprout, I took on the design of this project myself when we were down a team member, have shepherded it through development, and the feature is currently in beta testing.

• Design Manager: Idea champion, initially guiding project discovery and early design explorations, engaging with product leadership and stakeholders for buy-in
• Design Lead: Took ownership of project after ideation began, leading design, building interactive prototype, iterating, and defining MVP design specs
• Research Lead: Performed UX competitive evaluation, planned and facilitated usability testing
• Execution Partner: Embedded in the cross-functional team alongside developers to see feature through to delivery

Process & Methodology

Discovery Phase: User Pain Points Identification

Through customer feedback channels, ongoing generative user interviews, and watching session replays in Fullstory, we identified several recurring themes in how Sprout’s Compose experience created friction for our users:

• Customers found Sprout to be very efficient for writing and previewing how a single social post will appear when posted across multiple profiles and networks
• Social media managers often create a post for one network, then duplicate that post, tailoring the content for better engagement on each network
• Users found the “Schedule & Duplicate” workflow tedious and time-consuming

User feedback from various channels about cross-network posting in Sprout

Competitive Analysis

Finding screenshots and videos of enterprise social platform features is often difficult. Because Publishing is a core functionality of those systems, we were able to quickly gather insights about how most of our key competitors handle cross-network customization of social content. Without listing the specific competitor names or sharing the assets we reviewed, here’s a sample of what we found:

This analysis revealed opportunities to differentiate Sprout’s compose experience and ensure our customization solution was intuitive and competitive.

Ideation & Design

Early Concepts

With our problem clearly defined and some competitive examples to reference, it was time to practice some divergent thinking. We asked a lot of “what if” questions in the early stages of design for the project, like:

• What if we used generative AI to create the network-specific customizations for you?
• What if we showed customized posts as a “stack” of posts on our calendar views?
• What if we recommended that you customize existing cross-network posts for better performance?

As we started to converge on a scope of work we could realistically deliver on, I collaborated with the initial designer assigned to the project in exploring three distinct approaches:

1. Multiple content inputs – Simply duplicate the compose area for each selected network
2. Expandable panels – Multiple compose forms, with one shown at a time in collapsible section
3. Tab-based navigation – Horizontal tabs to switch between networks in the compose area

Early Figma concept with multiple inputs and a paper sketch of our first tab explorations

Design Evolution

After our first sync with engineering, we ruled out the feasibility of the multi-input approach. Our Compose experience supports scheduling, queuing, and sending over 30 post types across 9 social networks and has some of the most complex validation rules in the application. Keeping one version editable at a time was essential.

The accordion approach to expanding panels worked well, but identifying which network you were on was difficult. We were initially worried about horizontal real-estate with the tabbed approach, but when we switched to just network icons, there was plenty of room.

Dependencies & Scope Constraints

As we explored what it might look like to allow each network’s content to be edited independently, we identified a couple pre-reqs that needed to be tackled first:

1. Attached media and links needed to move out of their separate compose sections and into the editable content block
2. Our validation messages needed to change from a single list to being grouped by network

The platform engineers on our team had a backlog of critical projects, so we decided that the initial approach needed to be a primarily front-end project. The ability to keep customized posts grouped on the calendar and allow them to be edited together after saving would depend on database design changes and was therefore out of scope. We decided the user efficiency gain of a front-end-only solution was worth that sacrifice for our MVP release.

Prototyping & Testing

Interactive Prototype

With the team alignment on our general approach, I built a high-fidelity Figma prototype of the Customize Post concept to test with users, including:

• A guided post creation flow from initial compose to scheduling
• The ability to customize content and revert back to a cross-network post
• Attached post image shown in the editor block
• Scrolling the network preview to the currently selected network
• Validation messages grouped by network

Usability Testing

I decided that evaluative user testing of our proposed solution via targeted interviews would provide the qualitative input we needed to significantly de-risk the project. Based on Mixpanel usage data, I identified 25 customer accounts that frequently use our current Schedule & Duplicate flow from a variety of industries and company sizes. I used our customer-research channel workflow to request contact permission from each customer’s success manager and sent invites to sign up for a research session. My goal was 5-8 interviews, but I ended up facilitating 12 interviews over the course of 3 weeks.

Key insights from user interviews:

• Moving attached media – Most customers didn’t even notice. Those who did said it was a positive change and would expect attached videos and links to be there as well.
• Scrolling network preview – This was noticed and made sense to most users, but the scrolling was a distraction.
• Grouping errors by network – Made sense to everyone we spoke to.
• Red dot on X network tab – Almost everyone recognized this indicated an error with that network and most identified the issue (X was over character count) before even scrolling to the validation center.
• Network-specific compose blocks – We had Instagram Options shown for all tabs in the first few calls and everyone said it was confusing to be on the Facebook Tab but still see Instagram Options. We hid it when not on the Instagram tab for the remaining calls and only 1 customer thought that was confusing.
• Request: Customize other post details – A few customers expected to be able to change labels, campaigns, and other post meta data.
• Request: Customize send time – A few customers said they would send all variants at the same time, but most would prefer to use our Optimal Send Time feature to choose the best time for each network.

MVP Refinement

Based on our usability testing interviews, I made several design updates, which included:

1. Updating network preview column to show only the post for the tab you’re on
2. Designing an update to the scheduling UX to allow custom send times per network
3. Adding tooltips to red dot error indicators on tabs to show the error message

Implementation & Release Strategy

I met weekly with the engineering team through the development process, fielding questions that weren’t answered by the original kickoff documentation in Slack and participating in bug bashes as milestones became available to test.

Working closely with product and engineering stakeholders in a project kickoff, we planned a phased rollout approach:

1. Alpha release to Sprout’s social team and internal accounts
2. Beta testing with select enterprise customers, as well as our research participants
3. Limited release to 10% of user base with feedback collection
4. Iterative improvements based on initial usage data
5. Full release to General Availability with in-app education and onboarding

Expected Impact Metrics

While the feature is currently in beta, we’re expecting it to significantly change the way our customers compose cross-network posts, and ultimately their perception of Sprout. For this project to be a success, we’re measuring for and hoping to see:

• 60% reduction in “Schedule + Duplicate” actions (still helpful for some use cases)
• 45% reduction in session time spent customizing posts for multiple networks
• 30% reduction in cross-posting identical content to multiple networks
• 90% of beta users rating the feature as “very valuable” or “extremely valuable”
• Finally, we expect it to give us a bump in NPS rating and have a downstream impact on customer retention as well

Conclusion

The Customize Post Per Network project represents a significant evolution in Sprout Social’s compose experience. We transformed a pain point into a powerful capability that not only aligns with how sophisticated social media managers actually work, but sets us up for exciting, time-saving AI enhancements for our users. While I enjoy coaching designers through projects like this, by stepping beyond my management role to personally lead this project, I was able to shepherd it from concept through execution, ensuring that the solution truly met user needs while remaining technically feasible.

This project demonstrates my ability to:

• Identify high-impact product opportunities
• Lead comprehensive research to validate user needs
• Design complex workflows that simplify user tasks
• Partner effectively with cross-functional teams
• Balance immediate deliverables with long-term vision