Producing consistent, high-quality content at scale requires a system that empowers your team to write better copy faster. Ditto's AI content system turns your brand guidelines and copy standards into automated, actionable workflows — helping you draft, edit, and review copy so you deliver on-brand content consistently, from day one.
💡 More AI functionalities are coming (very) soon! If you have an AI related feature request or feedback, share it with us at support@dittowords.com or book a call
Use AI style guides to enforce your writing standards at scale
Build your style guide in Ditto with custom content standard rules, then let Ditto’s AI enforce them for you as you write. These rules provide examples to train your AI model and can be used to adjust tone of voice, use context-specific language, correct text formatting, and more.
Create a style guide:
Click + New style guide
Name your style guide and click Create
💡 You can set commonly used style guides to be enabled by default so it’s rules are applied in all projects
Create style guide rules to train your AI model
Style guide rules are the building blocks used to create your personalized AI content system. When adding new rules, provide at least one example to help the AI model understand the context and apply the rule correctly in your writing. The more specific you are and the more examples you provide, the better the suggestions will be.
Create a rule:
Click + Add rule in your style guide
Add a name, section, and description
[Optional] Add target tags so the AI knows which text items to apply this rule to. If no tags are included, the rule will be applied to all text items in your project
Provide at least one example, then click Add rule
💡 See here for more tips on how to write your rules so you get the best results from Ditto’s AI
Add to your word list
Ditto’s style guide has a dedicated Word List section that specifically manages your glossary of approved and disapproved terms. Use this to clearly list out the preferred terminology, words to avoid, and optionally note down the reasons for each use case.
Add a new word:
Click + Add word
Add the preferred word, alternatives to avoid, and notes, then click save
Enforce your content rules with Magic Edit
Ditto’s magic edit AI automatically checks your copy against the style guides enabled in your project, highlights any rule violations, and suggests appropriate edits. This provides you with immediate feedback on your writing and allows you to accept or reject the suggestions based on your judgment.
Respond to a suggested edit:
Select a text item and click suggested edits
See the rule that contribute to the suggestion and preview the edit
Accept (✓) or reject (✕) the suggestion. Accepting the suggestion will apply the edit to the text item and create the text item in your project
Use Magic Draft to write product copy in context
Magic Draft is your AI writing partner built directly into the Ditto. It writes product copy with full context into your Figma designs, style guide rules, and your entire Ditto workspace, so drafts are grounded from the very first suggestion, and get smarter as your team builds more context over time.
Instead of filling frames with lorem ipsums and writing copy up later, teams can use Magic Draft to jumpstart copy at the start of the design process and refine it collaboratively as work evolves.
Generate copy using Magic Draft:
Select one or more text layers, and click Magic Draft
Enter a prompt into the chat window the click enter. The more specific your prompt is, the better the result will be
Accept (✓) or reject (✕) the suggestion. The generated copy is displaced directly in your design so you can preview the results
💡 Magic Draft is limited to 20 text layers at a time. Select a subset of the text layers if you’re hitting the limit
Once you’ve generated your first draft, you have a few options to further refine the copy. Magic Draft is particularly effective for the functional copy that makes up the bulk of most products: error messages, tooltips, form labels, empty states, confirmation messages. Teams can generate these in bulk, review quickly, and focus energy on higher-value work.
Refine copy using Magic Draft:
Reject all: start the drafting process from scratch with the original copy
Refine: building off of the generated draft with a new prompt
Try again: returns to the last prompt and previous suggestions to have another go
How to get the best results from Magic Draft
Magic Draft draws from a hierarchy of context already in your Ditto workspace — your prompt, the text layer itself, Ditto metadata, the surrounding text in the frame and screen, and the broader content across your projects and component library. Drafts get more useful over time as your team adds more content and structure to Ditto.
A few things that make a meaningful difference in output quality:
Build your style guide before you start drafting: Rules are applied automatically, so the more complete your style guide is, the less manual refinement you'll need after the fact.
Keep text distinct within a frame: Magic Draft uses surrounding text to orient itself within the design. If a frame has duplicate copy, it can lose that context. Unique, descriptive text gives it a better starting point.
Iterate with prompts rather than restarting: Refine the result to steer the draft in a better direction rather than starting from scratch each time.
How to get the best results from Ditto’s AI
Due to the probabilistic nature of AI and large language models (LLMs), the output for AI suggestions can be variable in quality, and is partly influenced by how well your rules are written. Therefore, we want to share what we’ve found to yield the best results from Ditto’s AI so you can be confident in the suggestions it makes.
When writing dos/don’ts, ideally provide multiple, meaningfully differentiated examples to cover different use cases
Rules will be more successful if they are more concrete: "Don't say please" will be more predictable than "Use an assertive tone"
More subjective rules relating to voice, tone, and structure will generate good results, but they increase the likelihood of mediocre results
Names should be concise, direct instructions: "Use X instead of Y"
Descriptions provide context on how the rules should be used, and should also be thought of as direct instructions for the AI to follow
For vocabulary, formatting, and general best practices, it’s possible to only provide the do examples and skip the don’t
💡 Every Ditto workspace comes with a default Ditto Style Guide with common copy standard rules. You can use its rules as examples when creating your own custom rules