Your breakdown of the ROI really hits home. That "good enough polished clarity" for internal docs is the perfect sweet spot for the tool. I apply the same logic to data pipeline documentation - the AI polish gets our runbooks and connector notes to a clear, readable state quickly, which is a huge win for team onboarding. The human editor still gets the high-visibility customer-facing API guides, but the volume stuff? AI handles it.
Your point about the combined subscription cost is interesting. I wonder if that math starts to shift if you're only polishing technical docs and not doing broader content creation. For pure, dry technical writing, I sometimes wonder if a simpler, one-time-purchase desktop thesaurus might cover 70% of the need. But then again, the active voice nudge is the real time-saver, and I haven't found that in a traditional tool.
Do you find you're using it on more types of writing now that you've justified the cost, like emails or Slack announcements, or do you keep it strictly to docs?
ship it
Exactly my workflow. Scrivener for structuring GitLab CI/CD pipeline docs, then Wordtune for the final polish on the actual job descriptions.
Your dbt example is on point. I use it for turning dense Git merge strategy explanations into something a human can follow. The key is it keeps the technical steps intact.
Separating the tools is the only way I've found to scale. One for the messy architecture, one for the clean sentences.
Ship fast, review slower