Video Upload Flow

The challenge

The existing flow technically allowed users to upload multiple videos, but it behaved like a single-video editor.

After adding their files, users had to configure each video separately. The interface did not explain what information was required, which videos were ready or which still needed attention.

This created several usability problems:

• It was not clear that every video required its own title, description and settings.

• Users could not tell which videos had already been configured.

• Upload failures did not explain what went wrong.

• The interface did not provide a useful overview of a large batch.

• Users who wanted the same settings for several videos had to repeat the process one by one.

• Radio buttons next to each card suggested that only one video could be selected.

• The narrow panel layout left most of the workspace unused.

• Users did not understand that server upload and processing continued after completing the setup flow.

Because uploading is a key platform capability, this confusion affected one of the most important user journeys.

The design goal

The redesigned flow needed to work equally well for one video or a batch of dozens.

I defined four principles for the new experience:

• Show the entire batch clearly.

• Make each video’s readiness visible.

• Allow users to configure multiple videos safely.

• Clearly separate setup from server processing.

1. Turning the upload flow into a scalable workspace

The previous design focused on one video inside a side panel. This made the experience feel like a single-selection task and provided little visibility into the rest of the batch.

I replaced it with a responsive card grid that uses the available screen space.

All horizontal and vertical videos appear together in the same workspace. The grid adapts to the available width, including when the settings panel opens, so users can maintain an overview while editing.

This allows the same interface to support a single upload and a much larger batch without switching between separate modes.

From editing one video at a time to managing the complete upload in one workspace.

Responsive behaviour

2. Making the upload status of every video visible

In the old flow, users had to remember which videos they had already configured. There was no clear indication of what was ready and what still required attention.

I introduced visible states at the card level.

Each card communicates whether the video:

• Was uploaded successfully.

• Is missing required settings.

• Contains a validation error.

• Failed to upload.

Cards with missing mandatory information receive a clear red validation state. The corresponding fields are also highlighted inside the settings panel, connecting the problem to the place where it can be resolved.

Failed cards display an error description rather than a generic failure message. When the user continues, failed videos are automatically skipped and valid videos can still be processed. Users do not need to manually remove failed cards before completing the batch.

Users can understand the state of the entire batch without opening every video.

3. Clarifying title and description editing

The first step of the previous flow did not clearly communicate that users needed to review the title and description of every video.

In the redesigned experience, these details are part of the video card itself.

Users can click the edit action to update the title and description directly on the card. Clicking outside the editing area saves the changes automatically, reducing unnecessary confirmation actions.

When text is too long to display, hovering over the complete text area reveals the full value in a tooltip.

This makes editing feel like part of preparing each video rather than an unexplained step in a separate process.

Titles and descriptions are edited in context, directly alongside each video.

4. Applying settings to multiple videos

The biggest efficiency problem was repetitive configuration. Most users wanted to apply the same settings to several videos, but the old flow required them to configure every item separately.

I replaced the radio-button behavior with clear multi-selection.

Users can select any combination of videos and configure them together through a shared settings panel. This makes it possible to apply the same settings to an entire batch while still supporting individual changes when necessary.

Handling different existing values safely

Bulk editing introduced an important challenge: selected videos may already contain different settings.

Instead of displaying one misleading value, the panel communicates when the selection contains mixed values:

• Inputs display “Multiple values selected.”

• Radio buttons and toggles use a neutral state.

• Hovering over a neutral control explains that the selected videos contain different values.

• Optional values remain unchanged unless the user explicitly edits them.

If the selection includes both configured and unconfigured videos, a banner warns:

Some videos have no settings configured.

When the user first interacts with a setting that could overwrite existing values, a confirmation appears before anything changes. Once confirmed, the user can continue editing and changes are saved automatically.

This provides the efficiency of bulk editing without silently destroying settings that were already configured.

Existing settings are only overwritten after an explicit user action.

5. Preventing users from completing an unfinished batch

Users can select any combination of videos, but the flow must prevent incomplete videos from being processed accidentally.

If users click Done while mandatory settings are missing:

• Incomplete cards receive a visible validation border.

• Required fields are highlighted in the settings panel.

• Videos that are already configured keep their existing values.

• Optional fields do not block completion.

This makes the error actionable: users can immediately identify both which videos require attention and which settings are missing.

Validation appears both on the affected cards and inside the settings panel.

6. Separating video setup from background processing

Completing the video settings does not mean the videos are immediately available. They still need to upload and process on the server.

The old flow did not communicate this transition clearly, so users could leave without understanding what was happening.

I separated the experience into two distinct modes:

Focused setup

While users are adding and configuring videos, they remain inside a dedicated upload workspace. The global processing panel is hidden so it does not compete with the active task.

Background processing

Once the upload begins, users return to the video library and can continue working elsewhere. A global panel shows the status of the background upload and processing activity.

The panel remains visible while processing is active and closes automatically when all work is completed or dismissed. If another right-side editing drawer opens, it overlays the processing panel rather than resizing the active workflow.

Finishing the setup starts a visible background process—the user does not need to remain inside the upload flow.

The final outcome

The redesigned experience transformed video upload from a sequence of disconnected single-item actions into one scalable workspace for managing an entire batch.

Users can now:

• See all horizontal and vertical videos in one responsive grid.

• Understand which videos are ready, incomplete or failed.

• Edit titles and descriptions directly from each card.

• Apply shared settings to multiple videos.

• Understand and safely modify mixed settings.

• Identify missing mandatory information before continuing.

• Complete valid uploads without manually removing failed videos.

• Save unfinished batches as drafts.

• Continue working while upload and processing happen in the background.

Impact and next steps

The redesign addressed the main usability gaps in the existing upload experience and established a scalable interaction model for single and bulk uploads. It also defined the behavior for mixed settings, validation, upload failures, drafts and background processing—turning the feature into a more complete and predictable workflow.

To evaluate the redesign after release, I would measure:

• Successful upload completion rate.

• Time required to configure a multi-video batch.

• Number of uploads abandoned before completion.

• Frequency of missing-settings validation.

• Recovery rate after upload failure.

• Upload-related support requests.

Reflection

The main lesson from this project was that a bulk workflow cannot be designed as a single-item workflow repeated multiple times.

Users need visibility across the entire batch, efficient shared actions and safeguards for items that behave differently. Designing the states, validation and background behavior alongside the main screens helped turn a confusing upload form into a coherent system.