Research & Discovery

Existing user research

Use cases survey

We had breakdowns of the primary use cases of Paint users. Mainly consisting of Interstitial users, Productivity users and creative users.

Paint 3D post mortem

Paint 3D was a successor to Paint that saw poor adoption. In our study the key reason was it being too large a departure from Paint.

Competitive analysis

Competitive analysis was a huge boon for understanding art/creative app specific considerations, patterns, and how different apps had attempted to manage common constraints. Overall I ended up reviewing a selection of over 20 digital art apps across Windows, iOS, and Android.

Major Patterns

Most digital art/painting apps can be roughly broken down into two categories. Those that lean on patterns present in Adobe Photoshop and those that are closer to Procreate. Those with the Photoshop model tend to focus on customization and modularity as a key part of artist workflow. Those with the Procreate model tend to focus on painting as the core artist workflow and supporting that as a top priority with more advanced tooling hidden.

Above: An animated graphic comparing the influence of Procreate and Photoshop patterns

Opinionated simplicity

Modern apps focus on most used tools to limit what they surface to users. The interfaces are largely contextual in what they show. A digital painting flow is most supported and they limit the toolbar to a handful of core tools.

Procreate

Art Studio Pro

Infinite Painter

Above: Toolbar layouts from Procreate, Art Studio Pro, and Infinite Painter

Layout patterns

Modern apps mitigate or contain elements that block or limit the canvas surface area. Many trade offs are made in app design to keep the user in the flow.

I came to refer to this pattern internally as a windmill style layout. One interesting aspect to this is while similar layouts are used outside of digital art apps, there is an additional constraint that the content can not reflow in the same way, so we needed to be more rigid as far as the wider aspect ratios commonly used for digital art. With that it would be a consistent design consideration towards maximizing active canvas.

Research pillars

User testing

Typically run in conjunction with our user research team or ran by design using usertesting.com.

Feedback + Telemetry

We had a wealth of user-feedback and later into development telemetry that we could leverage.

Competitive analysis

A great resource for grounding research, and understanding dynamics of less testable scenarios.

Research & Discovery

Exploration

Blue-sky explorations

A large part of my process involved doing Blue-sky explorations of Paint as a more capable creative app. I would explore scenarios like modernized layout, improvements to shapes, color, transform tools, and text.

Research & Discovery

Features

Creator opportunities

With our purview we defined a few areas where we could build features with creative focused value. Layers, Brushes, and Zoom.

Layer updates

Context

Layers were one of the most-requested features, seen as a game-changer for expanding Paint’s creative capabilities. They create a clear delineation between Paint’s historical simplicity and its future potential as a full-fledged creative tool.

Layer Visual Hierarchy

We deliberated between a traditional list view and a more visual layer tile presentation. I advocated for using layer tiles because thumbnails naturally convey important visual information that might be lost with just text, especially when names are generic or overlapping.

Scope challenges

Our initial vision for layers included advanced functionalities (clipping masks, groups, blend modes, vector layers, opacity settings, locking mechanisms, and naming). However, due to a major rescoping and time constraints for our September deadline, the feature shipped without non-destructive saving.

Fast follow update

After launch, we discovered that introducing transparency affected existing workflows. In response, we quickly iterated and released a fast follow update: the addition of a background layer tile. This tile, present by default, can be hidden or have its color set independently, thereby ensuring the eraser and selection tool behave in line with user expectations. (Note: The implementation anchored the tile to the bottom rather than integrating it directly into the stack as originally designed.)

Brush updates

One area for improvement was brushes, while that was a large area. User feedback highlighted a specific point of frustration with size selection in the brush tool:

• Limited sizes: Only four fixed pixel sizes were available
• in-accessible: The brush size control was non-persistent, forcing users to reopen the size selector each time.

New approach

We revamped the brush tool to include:

• A refined slider control Offers a larger range of sizes while giving precision in smaller sizes and wider increments in larger sizes, adapting to the natural sensitivity required at different scales.
• A preview indicator Offers immediate feedback on brush size in the context of the image.

Zoom updates

The existing zoom implementation was limited only fixed 50% increments were allowed when the user zoomed in and out. The keyboard shortcuts were also very non-standard.

Implimented zoom updates include:

• Centering the canvas for more intuitive zooming and panning.
• Expanding the maximum zoom range.
• Introducing a precise combo-box control for exact zoom values, complemented by a fly-out menu
• A slider optimized for increased granularity, especially at higher magnification levels.
• A Fit to screen button to quickly return to the full view.

Zoom interactions

Zoom is interesting in that there are several common interaction patterns that can be considered standard. With that we wanted to offer a wide coverage of options.