Loading...
Loading...
Generic AI says “improve your course.” Hiring an instructional designer takes 6 weeks and $4,000. Course platform analytics tell you where students drop off, not why. Curriculum Auditor pastes your outline, runs a 6-dimension pedagogical audit grounded in Bloom’s Taxonomy, Cognitive Load Theory, and spaced retrieval research, and gives you module-level fixes in under 60 seconds.
No credit card. First audit free, lifetime. Average run: 47 seconds.
Every finding names the module and gives you the exact rewrite.
Not vibes. Scoring rubric is public and auditable.
Regenerates your full improved outline — ready to paste back into Teachable, Kajabi, Thinkific, Podia, or Skool.
Below is one real finding from an audit of an SEO course. Every audit produces 12–30 of these, grouped by dimension, sorted by severity. No signup needed to read this.
Module 2 introduces ~8 new concepts (keyword research, on-page SEO, technical SEO, link building, local SEO, content clusters, schema markup, site speed). Surrounding modules average 2–3. This is a textbook concept dump and a documented drop-off point in Cognitive Load Theory: working memory caps out around 4 novel concepts before retention collapses.
Fix:
Split into 3 focused modules: (1) Keyword Research & On-Page SEO, (2) Technical SEO & Site Speed, (3) Link Building & Content Strategy. Each becomes manageable in one sitting and the drop-off point disappears. Estimated implementation time: 2 hours.
concept_count: 8 · recommended_max: 3 · Bloom level: Apply · severity: critical
Audit findings are this specific because the model is forced to count observable features (concepts per module, modules with objectives, modules with assessments) before assigning a score — so the same outline submitted twice scores within ±3 points.
A PDF you can hand to a contractor, paste into a Notion doc, or print and scribble on. Six scored dimensions, two to five specific findings per dimension, and a Top 3 Quick Wins list at the top so you know what to fix first.
Each course outline is scored against six dimensions grounded in established instructional design frameworks. Weights reflect impact on student completion rates.
Bloom’s Taxonomy
Gagné’s Nine Events
Cognitive Load Theory
Spaced Retrieval Practice
Self-Determination Theory
ADDIE
Most “course improvement” tools either grade your marketing copy, summarise your videos, or hand you a generic checklist. We do one thing the rest don’t: a deterministic, module-level pedagogical audit grounded in published instructional design research.
| What you get | Generic AI course graders | Hiring an instructional designer | Platform analytics (Teachable / Kajabi / Thinkific) | Curriculum Auditor |
|---|---|---|---|---|
| Names the exact module to fix | No | Yes | No | Yes |
| Cites instructional design frameworks (Bloom, Gagné, CLT) | No | Yes | No | Yes |
| Same outline twice = same score (±3 pts) | No | Sometimes | N/A | Yes |
| Public, auditable scoring rubric | No | No | No | Yes |
| Turnaround time | Minutes | 2–6 weeks | Real-time | Under 60 sec |
| Cost for one audit | $0–$30 | $1,500–$6,000 | Bundled | Free first audit |
| Regenerated improved outline | No | Yes | No | Yes (paid) |
| Works on any topic (not just tech / business) | Varies | Yes | N/A | Yes |
Other tools say “add more quizzes.” We say: “Module 4 has no assessment — add a hands-on exercise where students build their first sales funnel.” You can act on it in one sitting.
The scoring engine runs at temperature 0 against a fixed rubric. The same outline submitted twice scores within ±3 points. Generic chatbots give you a different opinion every time you ask.
Paste a grocery list and we’ll politely tell you it isn’t a course outline — instead of fabricating a fake audit to look smart. The bar for a finding is “a designer would write this”, not “the model had something to say.”
Want a deeper comparison or to see a sample audit on your own outline before signing up? Email coursecurriculumauditor@gmail.com.
Run a complete audit on any course you’ve published, no credit card. If the report isn’t sharper than what you’d get from a $200 consultation, you’ve lost nothing.
If your first paid month doesn’t produce at least one finding worth implementing, email us within 30 days and we refund the $19 in full. No support ticket gauntlet.
The whole 6-dimension rubric is documented on this page. The model is forced to count observable features before scoring. You can disagree with a finding, but you can always see why it scored that way.
Most course-creator tools grade the wrong layer. They critique your sales page, summarise your videos, or score your “hook” — everything except the actual learning architecture. Meanwhile, every instructional designer worth hiring uses the same handful of frameworks (Bloom, Gagné, Cognitive Load Theory, spaced retrieval, ADDIE) to spot the structural reasons students drop off. Those frameworks aren’t secret. They just aren’t cheap to apply — until now.
Curriculum Auditor exists for the solopreneur creator with one to three published courses, a completion rate they’re not proud of, and no $4,000 to drop on a six-week ID engagement. Paste your outline, get a designer-quality audit in 60 seconds, fix the three modules that matter most. That’s the whole product.
Built by a solo founder. coursecurriculumauditor@gmail.com — replies the same day.
ChatGPT will give you a different review every time you ask, and it doesn’t cite frameworks. Our audit runs at temperature 0 against a fixed 6-dimension rubric — the same outline submitted twice scores within ±3 points, every finding cites the framework it comes from (Bloom, Gagné, CLT, etc.), and a pre-flight check rejects non-course inputs instead of fabricating an audit to look smart. The whole rubric is on this page so you can verify it yourself.
Yes. The rubric scores structural quality — learning objectives, module sequencing, cognitive load, assessment alignment, engagement design, completeness — and those apply across every domain. We’ve audited courses on copywriting, real estate, calligraphy, coding, and yoga so far. If your topic is unusual the model uses the source outline’s vocabulary in its findings (so “Chapter 3” stays “Chapter 3,” not “Module 3”).
Your outline is sent to Anthropic’s Claude API for analysis. We don’t train models. Anthropic’s API by default does not use your inputs to train their models either. Your audit row, including the raw outline, is stored in our database under Row Level Security — only you can read your own audits. Paid users can delete any audit at any time from the dashboard.
Course creators iterate. The single highest-leverage use of this tool is “audit, fix, re-audit, see the score move.” A subscription means you can re-audit the same course every time you ship a new module, plus audit additional courses as you launch them. If you only need one audit, the free tier is genuinely free, lifetime — no credit card.
Outlines under 30,000 characters are analyzed in full. Outlines between 30,000 and 50,000 are truncated to the first 30,000 with a visible warning, so you know exactly how much was analyzed. Outlines over 50,000 are rejected with a suggestion to split — an outline that long is usually two or three courses anyway.
No, and we’re explicit about that. A real ID engagement is multiple weeks of conversation, learner research, content development, and iteration — not a single audit. What this replaces is the “I just need someone to look at my outline and tell me what’s wrong” conversation, which is the part of an ID engagement that a $4,000 invoice mostly bills for.
Real testimonials land here once we’ve served the first 10 paying users. We don’t fabricate quotes — if you’d be willing to leave one, email us and we’ll send you a free month.
One free audit, lifetime. No credit card. 30-day money-back guarantee on paid.