UCSD, DSGN 1
2025
Instagram Post Feature Redesign
A research-driven redesign of Instagram's posting feature, informed by 18 user interviews and a competitive UI analysis.
Overview
Fixing the feature 78% of users struggled to complete
Instagram's posting feature is core to the app, but our research showed a huge number of users, even experienced ones, struggled to complete basic parts of it. I worked with a 6-person team to run structured user interviews, diagnose why posting was failing people, and redesign the three biggest friction points.
Problem
What was wrong with Instagram's posting feature?
We interviewed 18 users — a mix of frequent posters and people who rarely or never post — and had each of them walk through Instagram's full posting flow using a test account, narrating their steps out loud.
The results were clear:
16 of 18 users failed to draft a post. 12 couldn't do it at all, even after multiple attempts.
All 18 users hit friction with at least one core feature — most often multi-photo selection and the "advanced settings" menu.
Errors fell into two types: knowledge-based mistakes (users didn't know a feature existed) and description-similarity slips (users picked the wrong icon because two features looked alike).
The root cause wasn't a lack of features, as Instagram scored high on feature variety in our competitive analysis. It was the its weak signifiers:
Icons with no labels
Functions buried in settings
No visual cue for how to save or preview a post
Research
Our user testing showcased issues, especially in the "Draft" feature
The success rate for each activity conducted
To understand why posting wasn't working for everyone, we interviewed 18 users — a mix of frequent posters and people who rarely or never post — using the Master-Apprentice model, where each user walked us through their own process uninterrupted rather than answering leading questions.
Each interview had three parts:
Pre-activity — background questions on Instagram familiarity and posting habits
Activity — users completed a full posting flow (photo selection → filters/music → captions/tags/location → privacy settings → draft → post) on a test account, so no one had to post to their real profile
Post-activity — reflection on what felt hard, what they liked, and what they'd change
We logged successes, multi-attempt successes, and failures for every step, then coded each failure by root cause — separating knowledge-based mistakes (the user didn't know a feature existed) from description-similarity slips (the user picked the wrong feature because two icons looked alike).
Ideation
Redesign Prototype
Lo-fi sketch
Before landing on Instagram, we brainstormed broadly across digital and physical products with known usability issues, mapping ideas out visually to find the strongest opportunity. We chose Instagram's posting feature because of the clear split between users who post and users who don't.
Once research pointed us toward drafting, discoverability, and clutter as the core problems, we grounded our direction with a competitive design-space analysis, plotting Instagram against Tumblr, LinkedIn, X, and Facebook on two axes: feature variety vs. UI cleanliness, and signifier complexity vs. UI cleanliness. This showed Instagram had plenty of features but scored poorly on clarity — while LinkedIn (clean, low-feature) and Tumblr (feature-rich, cluttered) marked the tradeoff we were designing between. Our goal became a redesign that kept Instagram's feature depth while pushing cleanliness and signifier clarity toward LinkedIn's level.
From there, we sketched low-fidelity wireframes to quickly test different layout options — icon placement, labeling, and where to surface drafting and preview — before committing to final screens, following a double-diamond process (discover → define → develop → deliver).
Designs
Our Instagram posting feature re-design
Multi-Photo Select Button
The original multi-photo button was two unlabeled stacked squares — 5 of 18 interviewees either couldn't find it or found it only after multiple attempts.
I redesigned it as a photo icon with a plus sign, combining two well-recognized visual conventions into one clearer signifier for "add more photos."
Tradeoff: a slight learning curve for users already familiar with the original icon.
Photo Customization Layout
The original layout scattered customization options inconsistently. Music sat in its own browsing area below the photo, while filters, text, and image overlays were unlabeled icons grouped separately at the top right.
I consolidated everything into one consistent, labeled toolbar, giving each feature a clear icon + label pairing for better discoverability.
Tradeoff: added labels increase clarity but risk visual clutter, and long-time users may need to adjust to the new placement.
Post Screen Setting
The original settings screen was cluttered and repetitive — music editing was duplicated from the previous screen, and key features like audience control were buried without clear labels.
I redesigned it with the same labeled-icon system as the customization screen, then made two targeted additions based on our biggest research findings:
Draft this post — only 2 of 18 users could figure out how to draft (it previously required exiting the flow entirely). Drafting is now a one-tap option.
Private Preview — several interviewees cited anxiety about how a post would look to others as a reason they avoided posting. A new eye icon lets users preview their post in-feed before sharing.
I also replaced the single "Share" button with two clearer options — "Your posts" and "Close friends" — to remove the ambiguity of the old audience settings.
Tradeoffs: simplifying with labels risks clutter if overused; drafting adds flexibility but users may accumulate forgotten drafts; and preview, while reducing anxiety, could slow down posting or add pressure to "perfect" a post before sharing.
Lessons
What I learned
Invisibility, not bad design, was the real issue. Instagram already had the features people needed. They just weren't signaled clearly enough to find. Sometimes nothing's broken, it's just hidden.
"Users struggled" doesn't always mean the same problem. Some people didn't know a feature existed. Others knew it existed but picked the wrong icon because two looked too similar. Different problems need different fixes.
Every fix came with a tradeoff. More labels meant more clarity but also more clutter. A draft option helped but risked drafts piling up unused. There was no version of this redesign without downsides, just versions we could live with.
Letting data lead beat relying on instinct. Drafting wasn't the problem I expected to be biggest. The interviews made that clear, not assumption.
Not every barrier was a UI problem. Some people avoided posting out of anxiety about how it would look to others, not confusion. That's what led to the preview feature.
