In content teams that lean on AI, thinking drops out first. The drafts read fine and say nothing. Most Heads of Content can feel it without being able to name what's wrong. That's exactly what I do.
These are not writing problems. They are thinking problems. The distinction matters because fixing the writing does not fix the thinking. Most content feedback addresses the symptom and leaves the root cause completely untouched.
Here is what the audit finds most often:
Every piece is scored against a four-part rubric. Every writer gets a named failure pattern. The team gets a map of where the thinking breaks down and the three structural fixes that will move the work fastest.
| Dimension | Writer A | Writer B | Writer C |
|---|---|---|---|
| Angle construction | 42 | 78 | 61 |
| Audience thinking | 38 | 82 | 44 |
| Evidence usage | 55 | 74 | 39 |
| Originality | 31 | 66 | 28 |
For a 6-writer team at the founding rate.
I have been reviewing content professionally for six years. For most of that time I was looking at the writing and missing the actual problem, which was happening before anyone opened a document.
When I started tracking thinking failures specifically, not prose quality, not SEO compliance, not tone, the patterns became visible almost immediately. The same four failures repeating across otherwise talented writers on otherwise well-briefed topics.
I built the rubric to make those failures nameable. Named failures get fixed. Vague feedback about "needing more depth" or "stronger angles" does not.
This is the system I use on my own team. It is the one I run on yours.
The eight-page document is the actual review system behind the audit. The taxonomy, the rubric, and a worked example with one real before-and-after. If you want to understand what thinking failures look like in practice before spending anything, read this first.
You can also run a version of this on your own team right now. The framework is the audit. The service is what you get when you want someone else to run it and tell you what they find.
The taxonomy. The rubric. The worked example. This is the actual system. Read it and you will immediately be able to apply it to your team's work.
Tell me how big your team is and send a few recent pieces. I will give you one specific observation for free, then tell you whether the audit fits. One message is enough.