Suggest a Feature

As a designer, I'm working on a large-scale Figma project with over 3 years of development. The file includes numerous deeply nested components used throughout the product.Even minor changes to a single component—such as swapping a fill for a stroke on a button—trigger severe performance problems:Figma freezes for 5–10 minutes as changes propagate across all instances. Another 5–7 minutes are needed to sync with the server. The tool becomes sluggish, with 1–3 second lag on every action due to background updates.This makes efficient work nearly impossible. I can't make consecutive changes without waiting for updates after each one, significantly hampering my workflow. Proposal:Implement a feature that allows users to temporarily freeze a component, preventing changes to the parent component from immediately affecting its instances.Implementation concept:Enable a "freeze" mode for the component. Make multiple edits without triggering updates to child instances. When finished, unfreeze th

I’m building a web app. The issue I have is that the mobile frame options in Figma don’t include the header and footer that browsers have. So when you preview your work, it looks like a native app, not how you’d actually see it when browsing the web on a phone. I know I can manually add this as a header and footer for every frame for every page of the site I’m designing, but it’s kind of a pain. Real mobile browsing includes these headers and footers of the browser. 2. 3: Would like to see an option to turn on this kind of overlay for frames, so I can easily design for mobile browsing, and not just for native apps.

Does anyone know how to turn off Figma’s auto-formatting? I’m trying to type '- ', but it looks like Figma is assuming this text must be a bulleted list. I’m having a similar issue when using text that’s in the format of an email address or a url. Figma is changing these to unwanted links. I’d like to turn off all text auto-formatting.

When creating tokens and working with variables Dev Mode generates code where the group name is included as part of the token name. Currently this doesn’t make too much sense (for me at least) and in Figmas own variables tutorial it would result in spacing tokens being called “SpacingSpacing-sm” if you would look at the tokens in Dev Mode. Can we please get an option to exclude the group names from the generated code? The groups as I seem them are for organising and not necessarily something we want to automatically include in the token name. Semantics are important for tokens so if I have to leave out part of the name and only have it as part of a group name it makes reading the token and using it correctly in Figma so much harder. Definitely interested in hearing others experience regarding this! Maybe i missed something or should rethink our structure.

The Problem: > Currently, Figma’s native PDF export converts all text layers into vector paths (SVG objects). This causes several major issues: High File Size: Vectorizing every character creates unnecessarily large files. No Accessibility: Text is not selectable, searchable, or readable by screen readers. Professional Limitation: We cannot use Figma for brochures, data sheets, or documents that require copyable text. Proposed Solution: > Implement Font Embedding (subsetting) in the PDF export engine. Retaining text as a "text layer" would make Figma a viable alternative to Illustrator or InDesign for document design.Why it matters: > This would significantly improve workflow for designers creating marketing assets and technical documents within Figma, ensuring files remain lightweight and fully functional

Hi all! The new Slot has just been released and I’m looking into replacing the selfmade slots in our library with the new released slots. But I’m already walking into my first problem with the new slots. I have a component where we have 2 slots with the same instance swap property. These slots have boolean variables added to them so they will be hidden/shown on different breakpoint modes.With the new slots I have not found a way to use the same slot property twice in the same component yet, other than keeping our current setup for now, does anyone have a good idea on how to solve this?

Title should read “Open specific page in Figma Make prototype”I have a Figma Make prototype that I am using in a navigation usability test. We are testing the global nav, and people’s understanding of where they land vs where they need to go. My prototype only opens on the default page, however I need the prototype to open on a variety of pages for my test. Is it possible to have it open somewhere other than the default?

The issue being faced today:I want to have preferred values only show up under the instance swap feature. Not searchable in the assets panel, as these are typically sub-components that I don’t want to be used on their own (examples: a dropdown menu used in a global navigation component, the vector for an icon used to instance swap in an actual icon component to control the sizes via icon properties, a single segment used inside a donut chart, pre-configured links or buttons commonly used inside a bigger component to make it easier to select from a pre-set). Why this isn’t possible:Unpublished and hidden components are treated the same in Figma. Hiding/unpublishing a component in Figma is currently the only way for it to not show up in the assets panel. Hiding/unpublishing a preferred value leads to additional issues: Preferred values cannot be selected by library consumers if they are unpublished. Preferred values must be published in order for updates to be available to library consu

Would Figma consider supporting Variables for variable font axes in the future?We're using Google Sans Flex and switch between two visual modes:Formal - `Rond = 0`Friendly/Cute - `Rond = 100`Example of the ROND (Round) axis in Google Sans Flex. We use this setting to switch between formal and friendly typography styles.​​​​​Currently, font axis values such as Round (ROND) cannot be controlled by Variables. Because of this, we have to maintain two separate text-style systems even though everything else is identical.after add rondIf variable font axes could be bound to Variables, we could switch typography modes through Variables and maintain a single text-style system instead.Example concept: Variable Font axes such as Weight and ROND could be connected to Figma Variables, allowing typography modes to switch automatically.This would be especially useful for design systems that use variable fonts to express different brand personalities across themes or modes. 

Is it possible to map the HUG or FILL to a variable. It would be great to not only assign particular width values, but also set the width to either hug or fill the container.

This may be a hacky way to achieve what should probably be a standalone text decoration style (highlight)… but with the addition of underline styling, you can basically setup a highlightThe issue is that the underline appears over the text. It would be great to have to option to set the position or z-index of the underline to appear behind the text? 

We currently manage over 800 templates across roughly 20 collections in Figma Buzz. When using the Grid view, the scrollable area limits out at approximately the top 50% of the loaded templates. To access templates at the bottom of the grid, we are forced to manually collapse the collections above them. This is an unnecessarily cumbersome workaround that slows down our workflow. We should be able to infinitely scroll or navigate smoothly from the very top collection down to the bottom collection in Grid view without hitting an artificial scroll limit or needing to collapse preceding folders.

Hot take,Figma Make, in it’s current beta form, is underwhelming when it comes to being able take a design and turn it into a usable coded prototype quickly.This feature will have a huge impact on designer-developer workflows only when the speed to getting to a Make prototype is faster than getting to a noodle prototype. Until that happens, designers will continue to stick with what is fastest/most effective for them.Being able to provide Make with the entire context of a prototype + developer notes + design library inside a given file/page seems to be the best route to doing this.I hope the refactor needed to get to being able to provide access to all of this data is well underway, but if not, I’m putting my trust in the rest of the community to get some movement behind this idea here in the portal.

There are many legacy companies starting to value design a lot more, and they are hiring designers into their ranks to make their software products a lot better. At these places, there may be strict network or security requirements that our critical software must be self-hosted or deployed on-premise. For us designers, we join these companies in hopes of showing that design is about collaboration and working together, not about opaque hand-offs and “design black boxes”. What better tool do we have than Figma, to show the collaborative nature of design? Please please please do allow for on-premise deployments of Figma — and help us all in traditional companies bring change slowly, one tool at a time. 🙏 I understand that this feature request is pretty far-off from Figma’s product philosophy so far, but well… 🤞 (Hey, our companies usually can pay too.)

Need this feature for icebreakers in workshop please.

When working on complex files with deeply nested layer structures, the fixed left sidebar creates a significant workspace constraint. On a single monitor, navigating the layer hierarchy competes directly with canvas space. On multi-monitor setups, there is no native way to move the layers panel to a second screen so the full canvas width can be used on the primary display.I would like to be able to detach the layers panel and open it as a standalone, resizable window, either on the same monitor or on a secondary display. This would allow me to keep the full canvas in view while navigating and inspecting the layer tree, particularly on large or structurally complex files where frequent layer panel interaction is required.This would benefit designers working on multi-monitor setups, those managing files with deep nesting or high layer counts, and anyone who finds the fixed sidebar a bottleneck during detailed inspection or handover work.

In Figma, all the boards, etc can only be sorted by when I last viewed. That’s not very useful (I know what I’ve viewed) What would be much more useful would be to view by last edited so I can view what is the most recently changed thing.

ProblemWhen a vertically scrollable frame is nested inside another vertically scrollable frame, scrolling works inside the child until it reaches its top or bottom edge — at which point the scroll locks in place. The remaining scroll input is never passed to the parent frame, so the user has to physically move their cursor outside the child frame to continue scrolling the page.Current behaviorScroll input is captured entirely by whichever scroll container the cursor is over. When that container hits its limit, the input stops there and does not propagate.Expected behaviorMatch how browsers handle nested scroll containers by default: once a child scroll region reaches its edge, the leftover scroll should "chain" up to the parent, letting the user keep scrolling the page in one continuous gesture without repositioning the cursor.Proposed solutionAdd a scroll-chaining option to the Overflow settings on the Prototype tab — equivalent to CSS overscroll-behavior. Ideally with two modes:Chain

Problem: When defining "Preferred instances" for the new native Slots, we can only select base Component Sets. Inserting one always loads the default variant, requiring extra clicks to reach the desired state.Request: Please allow us to pin specific variants (e.g., State = Active) directly in the preferred instances list.Why: Saves clicks, provides strict guardrails, and clarifies exactly which variant is intended for that slot right out of the box.

I see you’ve added the ability to name sections now in the overview view in slides, big fan. But what i thought would be cool would be the ability to add that title/text as a dynamic piece of text to a slide like a page number.This would be really useful for Overlines/footers in templates and live documents where you might want to add signposting to a slide. For example my slide footer will have the section name on the left, and rather than having to edit this on ever slide, having it auto populate from the section name would be awesome.

Right now, once you create a Main Component, there is no way to convert it back into a normal Frame/Group/Autolayout. If a component is no longer useful, we are forced to use the workaround: create instance → detach → delete main component. This adds extra steps and slows down workflow. Request: Please add a simple option like “Remove Component” that instantly converts a main component back to its original state.This would save time, reduce friction, and improve cleanup when working with experimental or temporary components.

Hi there!We use the comment feature (C) a lot in Figma for feedback and collaboration across the team, but it doesn’t work well for tracking design tasks or to-dos.Right now we misuse comments for task tracking, which creates clutter and causes important items to get lost. Also, not all tasks should be visible to the wider team outside design team.What we’re missing is a simple task system for designers! Nothing heavy — just a clean way to track internal design tasks without mixing them into comments.Here an example 🙂TaskFor example a ‘Task tool’ similar to a comment tool with features like:text field assign to @name mark as done optional priority (high) This would significantly improve workflow for design teams. Thanks! 🙏

Hi there,One issue we noticed with the new Slot feature is discoverability inside components such as tiles or cards.Currently, when a slot is configured in the properties panel but the final component is not hovered, the slot area is completely invisible. Because of this, users do not realize that there is an adjustable slot inside the component.At the moment, the slot only becomes visible on hover, which makes the feature difficult to discover. I attached screenshots showing:a non-hovered state where the slot is invisible   a hovered state where the slot becomes visible   My concern is that slots should remain somehow visible when empty.Would it be possible to introduce a subtle always-visible indicator for slots, without requiring a full background color? 

Hi all,First of all, the new Slots feature is truly awesome, I can’t wait to see how in depth we’re going to rework the components of our design system with it. I’d like to suggest a small improvement regarding their visibility. When a slot has no fill or background, it’s basically invisible on the canvas unless you hover over it. This makes it harder to understand the structure of a component at a glance, especially for our designers who are going to consume the components we provide.It would be really helpful if slots could have a dashed outline, similar to the dotted boundary used for variant groups in component sets, so empty slots remain visible without adding a background.I tried looking through the settings to see if this was already an option. If it exists and I simply missed it, my bad. But if not, having something like a default or optional slot outline could make components much easier to read and work with.

Hello, I would like to propose an improvement for navigation in the design files.In our project, we add many internal links to specifications details, which facilitates the timely consultation of information.On the other hand, after following an internal link, it lacks a simple way to return to the previous location in the file, such as a floating Back button or a behavior similar to "Return to instance" after navigating to a component definition.It would also be very useful to be able to automatically display the list of backlinks pointing to a given section, frame, or element.This type of functionality would greatly improve navigation, reading documentation, and working with complex files that have a lot of cross-referencing.