Do Not Panel

A do not panel is a visual callout in your brand book that displays realistic examples of off brand work with sharp annotations that show exactly where and how the mistake breaks your rules. It sits alongside your positive guidance and serves as the clearest way to train eyes and instincts across design marketing and AI assisted workflows. The 2026 brand identity guidelines article nailed it when it said this is the section almost every brand book skips yet the one people actually remember and reference because nothing cements a rule like seeing the ugly consequences of ignoring it. These panels use real looking production artifacts not theoretical diagrams and they call out violations in logo placement color semantics motion timing voice tone AI output and social formats with specific labels that name the sin and the impact like increased bounce rates or brand recognition failure in user tests.

It is not a bullet point list of things to avoid written in corporate euphemisms. It is not a collection of abstract red X graphics over stock images like those 2019 Webflow PDFs that collected dust on shared drives. It is not a soft reminder or an appendix you add at the last minute to hit page count. It is not vague or open to interpretation. It definitely is not something an LLM can parse if you only give it adjectives instead of these concrete bad examples paired with clear corrections.

Take the concrete example from Slack's 2024 brand system overhaul. Their do not panel for iconography showed the correct simple multicolored emoji style icon next to a version where a vendor had used overly detailed illustrations with gradients and shadows. Red arrows pointed to the extra complexity the color clashes with their semantic token system and the way it broke consistency across their 16 by 9 webinar slides. The caption read This turns our friendly workspace into a cluttered mess that increases cognitive load and reduces user satisfaction scores by 18 percent in testing. In the voice section they displayed an on brand notification Your channel has new messages versus the off brand Your designated communication hub has received fresh correspondence with annotations calling out the stiff formality that contradicts their approachable casual personality. They added a panel for motion that contrasted their quick 200 millisecond ease out on hover states with a bouncy overshot animation labeled Do not add playfulness that feels gimmicky and distracts from the content. These panels were built directly in Figma as components so teams could drag them into their own files during critiques and they got referenced in over 300 design critiques in the first quarter after launch.

Expand that thinking to a company like Dropbox in 2025. After their AI integration push they created do not panels for image generation that showed a prompt following their positive style direction of clean minimal workspaces next to one that ignored the negative prompt list and produced clichéd stock office photos with smiling diverse teams staring at screens. The annotation listed the banned elements like floating heads lens flares and the wrong color temperature that made it feel like every other tech brand instead of distinctly Dropbox. Their typography panel showed correct use of their custom font at proper sizes and weights versus a version using system fallbacks that created hierarchy collapse on mobile views with measurements tied back to their design tokens. One particularly brutal panel tackled the exact phrase Lets synergize our growth journey as an example of voice violation that sounds like a bad LinkedIn post instead of the direct human language their rules demanded. Shopify took a similar approach in their 2023 merchant guidelines with a do not panel for partnership co branding that showed a correct balanced logo lockup versus one where the partner mark overpowered the Shopify icon leading to trademark complaints and diluted platform authority. Notion's 2025 refresh added an accessibility panel with a button using primitive gray 300 on white labeled Do not ship this 2.9 contrast ratio fails WCAG AA and locks out 6 percent of users.

When to use a do not panel is every time you define something that could be expressed incorrectly in visual or written form. Roll them out in the foundation section for personality attributes by showing we are quirky but we are not childish with side by side mood boards. Use them heavily in visual identity for logo do nots like placing it on busy patterns or distorting proportions. They become lifesavers in the motion chapter where a wrong easing curve can make your entire product feel cheap and in accessibility by showing a button with 2.8 to 1 contrast ratio labeled Do not ship this it fails WCAG AA and excludes users with low vision. Deploy them in voice rules with full sentence pairs and in the AI section where they make negative prompts tangible by showing the exact bad output you want to avoid from tools like Midjourney v6 or Flux.1. They work for social first variants by showing a hero image composed for 9 to 16 that maintains legibility versus one that loses all text in the crop leading to 35 percent fewer impressions. Bring them into onboarding decks vendor contracts freelance kickoffs and even your CI pipeline test cases so automated checks have visual references. Do not use them when the rule is purely technical like how to structure your JSON token file or who approves version changes in governance because those do not benefit from visual edge cases. Avoid them if your team lacks the skill to make the bad examples look convincingly like real work or if showing certain violations could create IP or legal headaches with past campaigns.

The secret to powerful do not panels lies in the details. Make the bad examples slightly too realistic so they trigger instant recognition from anyone who has worked on the brand. Use the exact same annotation style across the entire book with your brand red at full opacity not some soft pastel. Keep labels under nine words. Always tie the violation to a real outcome whether it is lower conversion rates confused customers brand dilution that takes months to recover from or support tickets that spike after a bad release. Update them quarterly as new mistakes appear in the wild from your distributed team or from the latest AI model hallucinating your style in unexpected ways. Test the panels with new hires to confirm they can identify the problem without reading the labels first.

A do not panel turns abstract brand rules into immediate visceral understanding that sticks with your team long after they close the book.