LoginGet started

Generation · Content editing

Improve Drafts: Preserve Voice, Structure & SEO

Apply targeted edits to existing copy—clarify language, fix structure, optimize for keywords, and keep original formatting for seamless CMS export. Designed for editors, SEOs, and teams managing many pages.

Try Content ImproverSee prompt templates

Workflow

Editor‑first

Focused on improving existing copy (clarity, structure, SEO) rather than generating new content

Formats

Docs, Markdown, CMS

Accepts Google Docs, .docx, Markdown, and CSV for batch edits

Output

CMS‑ready

Preserves HTML/Markdown structure for direct export or pasteback

Solve common editorial pain points

Why use a content improver

Editors and SEO teams need a tool that shortens rewrite cycles without losing the author's voice or structure. The Content Improver addresses unfinished drafts, thin SEO content, and time-consuming manual QA by offering guided edits, SEO-aware rewrites, and export-safe output.

  • Turn underperforming pages into search-ready content without full rewrites
  • Keep tone consistent across authors with granular tone and length controls
  • Reduce manual QA by presenting suggested edits as diffs that are quick to review

Editor-first workflow

How it works

Load a single page or a batch, pick an editorial prompt cluster, tune controls (tone, reading level, target length, target keywords), then preview suggested edits side-by-side with the original. Accept, reject, or tweak each suggestion and export as HTML or Markdown for your CMS.

  • Preserve structure: headings, lists, links, code blocks and inline HTML are retained
  • Granular controls let you adjust sentence length, tone, and reading level per section
  • Audit-friendly diffs show each change with a rationale tag (SEO, Clarity, Tone, Length)

Preserve formatting

Edits are applied to the content while keeping original markup and structure, minimizing copy/paste rework.

  • Keeps H1–H6, lists, tables and code blocks intact
  • Outputs valid Markdown or HTML-ready snippets

Granular editorial controls

Control tone, length, and reading level per document or per section to keep brand voice consistent.

  • Shorten or expand sections by percentage
  • Select target tones (e.g., technical, conversational, formal)

Side-by-side diffs & rollbacks

Suggested edits appear next to the original with one-sentence rationales and tags for easy review.

  • Tag edits as 'SEO', 'Clarity', 'Tone' or 'Length'
  • Accept, reject, or modify suggestions inline

Editorial templates tuned for real tasks

Prompt clusters & practical examples

Use prebuilt prompt clusters that target common editorial tasks: clarity, SEO, headlines, summaries, localization and accessibility. Each cluster contains conservative instructions so the tool edits text without inventing facts.

  • Improve clarity and flow: "Rewrite the following paragraph to be clearer and more concise while preserving the author’s voice. Keep technical terms intact. Output only the edited paragraph.\n\nOriginal: {content}"
  • SEO optimization: "Given this page content and these target keywords: {keywords}, suggest a rewritten page intro, new H2s optimized for search intent, and a 150–160 character meta description. Preserve existing facts.\n\nContent: {content}"
  • Change-diff and review: "List suggested edits as a side-by-side diff with a one-sentence rationale for each change, and tag each edit as 'SEO', 'Clarity', 'Tone', or 'Length'.\n\nContent: {content}"

Work where your content lives

Sources and export formats

Import content from common editorial systems, run edits in-place, and export back in CMS-ready formats. The tool is designed to slot into existing editorial and deployment workflows without forcing a complete platform migration.

  • Import from Google Docs and exported .docx files for single-page edits
  • Roundtrip Markdown or HTML for headless CMS and static-site repos
  • Batch process CSV or Google Sheets columns for large sets of pages

Scale editorial rules across many pages

Batch workflows & governance

Apply consistent editorial rules to hundreds of pages with CSV-driven templates and rule sets. Use staged review to route changed pages to reviewers and preserve an audit trail for compliance and version control.

  • Bulk-edit template: apply H1 checks, duplicate paragraph removal, and meta description suggestions to CSV rows
  • Staged approval: preview changes, assign reviewers, and export accepted edits
  • Export suggested edits with row IDs for import into Git or CMS workflows

Safe edits for teams

Review, audit and export

Suggested edits include short rationales and tags. Editors can accept changes individually or in bulk, roll back edits, and export final content with the original structure preserved so QA and legal review remain straightforward.

  • Diff view with rationale and change tags
  • Export accepted edits as Markdown, HTML or a CSV with original row IDs
  • Short changelogs for accessibility fixes and passive voice corrections

FAQ

How does the content improver preserve the original author’s voice while making edits?

The workflow maintains the original text and applies targeted edits rather than replacing content wholesale. Controls for tone, reading level and edit aggressiveness allow you to prefer light copyediting or stronger rewrites. Side-by-side diffs show each edit so humans can accept only changes that preserve voice.

Can I optimize an existing page for a new set of keywords without rewriting facts?

Yes. Use the SEO prompt cluster to suggest a rewritten intro, new H2s and a meta description while instructing the model to preserve factual claims. Suggestions are presented as edits—reviewers can accept only framing and headline changes while leaving factual sentences untouched.

What formats and content sources can be batch-processed?

Common sources include Google Docs and exported .docx files, Markdown files from repos or headless CMS, Notion pages, and CSV or Google Sheets for bulk runs. The tool preserves markup where possible for direct export back to your publishing workflow.

How should editors review suggested edits to avoid accidental fact changes or hallucinations?

Reviewers should use the side-by-side diff, check the rationale tags (SEO, Clarity, Tone, Length), and verify any factual changes against source material. For bulk runs, start with a small sample batch, confirm no factual drift, then scale up with conservative edit aggressiveness.

Does the improver maintain HTML/Markdown structure when exporting back to CMS?

Yes. Suggested edits are applied in ways that preserve headings, lists, links and code blocks so exports are CMS-friendly and require minimal reformatting.

What controls exist for tone, reading level, and length?

You can set tone (e.g., conversational, formal, technical), target reading level, and percent-based shorten/expand rules at the document or section level. These settings help maintain brand voice and keep changes predictable.

How does the tool help produce meta descriptions, social summaries, and improved headings quickly?

Use dedicated prompt clusters for microcopy: generate multiple H1/H2 suggestions, a 150–160 character meta description, and 25–35 word featured-snippet summaries. Results are returned as discrete suggestions so editors can choose or iterate.

What is the recommended workflow for bulk-improving low-performing pages?

1) Export page bodies and IDs to CSV, 2) apply bulk-edit template with desired rules and target keywords, 3) review suggested edits on a representative sample, 4) approve and export accepted edits for CMS import, and 5) monitor performance and iterate.

How do localization and regional adaptations work?

Use the localization prompt cluster and specify the target locale. The tool adjusts spelling, units, examples and tone while preserving legal disclaimers and core facts. Localized suggestions remain editable and are returned with rationale for quick review.

Can suggested edits be exported for version control or audit purposes?

Yes. Export options include Markdown or HTML snippets and CSVs that map suggested edits back to original row IDs with changelogs and rationale tags to support version control and audit trails.

Related pages