Close the loop. Let AI read the patterns.
Why this mattersMost business owners create content, publish it, and never look back. They have no idea which posts moved direction requests and which ones did nothing — so next month they post the same kind of content and get the same kind of results. The Performance Loop closes that gap in 10 minutes a month, on autopilot.
Read this once. Sit with it before you answer.
How do you turn last month's analytics into next month's content plan in under 10 minutes, without doing the pattern recognition yourself?
Where you stand right now.
By the end of this lesson, you will be able to:
- 1Run the Content Performance Loop Prompt against your current 30 days of GBP and social data.
- 2Provide the three required inputs (GBP Performance data, social analytics, published post list with types).
- 3Read the five-section output and identify what to keep, what to cut, and what to add for next month.
- 4Save the output and apply the three recommendations to next month's cadence.
The whole lesson in a few points.
- 01Lesson 8.3 is the manual ritual. This lesson is the upgrade — AI does the pattern reading.
- 02Three inputs: GBP Performance data, social analytics, list of posts with types.
- 03Five output sections: which topics drove activity, which post types performed, what your audience wants more of, what to stop, three specific recommendations.
- 04Save the output. Apply the recommendations next month. Repeat monthly.
- 05The compounding kicks in by month three. By month six your content is materially different and better targeted than what you would have produced without the loop.
What the loop is actually doing.
The Content Performance Loop is the missing piece in most content systems. Most business owners create content, publish it, and never look back. They have no idea which posts moved direction requests and which ones did nothing — so next month they post the same kind of content and get the same kind of results. The loop never closes.
The Performance Loop Prompt closes the loop. It takes your data from the Performance tab, combines it with your social analytics if you have them, and cross-references against the posts you actually published. The output tells you three things: what content drove the most engagement, what patterns are emerging, and what to do differently next month.
This is not magic. The prompt is doing pattern recognition you could do manually if you had two hours and a spreadsheet. The prompt compresses that into 10 minutes by pulling the patterns automatically.
What data to paste in.
You need three inputs for the prompt to do its best work.
Input one. Your GBP Performance data — discovery searches, direct searches, customer actions including calls and direction requests, photo views, and the top search queries from the last 30 days. Screenshot the dashboard and describe the numbers in the prompt, or use the Looker Studio GBP connector if you have multi-location operations.
Input two. Your social analytics from any platform where you actively post — Facebook, Instagram, LinkedIn. Reach, engagement, profile visits, link clicks. If you only post on one platform, that is fine. Use what you have.
Input three. The list of posts you published this month with their topics and post types — education, social proof, conversion, human / BTS — from your Lesson 7.4 cadence worksheet. This is the part most loop attempts skip, and it is the part that makes the analysis actually useful.
You do not need to be exhaustive. The prompt works with partial data. It will tell you what additional context would improve the analysis if you want to iterate.
What the output looks like.
The prompt returns five sections. One: which content topics drove the most profile or website activity this month. Two:which post types performed best relative to reach — education, social proof, conversion, human. Did your audience respond more to one type than the others?
Three: what the data suggests your audience wants more of. This is the pattern read — the prompt looks for repeatable signals across your posts. Four: what you should stop doing based on the data. Sometimes a topic or format you have been running is not actually performing, and the prompt will flag it. Five: three specific recommendations for next month. Not generic advice — specific topics, post types, or angles based on what your particular data showed.
The output reads like a strategy memo from a marketing director, except it is grounded in your actual numbers, not best practices. Save the full output in your workbook on the Performance Loop Output page. You will use the three recommendations to plan next month's cadence.
How this changes your cadence over time.
Month one, you run the prompt and get three recommendations. You apply them in month two's cadence. Month two, you run the prompt again — the data now includes the impact of last month's recommendations, and the new recommendations are sharper because they are calibrated to your specific audience response.
Month three, the patterns become clearer. You can see what is working and double down. You can see what is not and cut it. By month six, your content is materially different from what you would have produced without the loop — more specific to your audience, more aligned with what actually moves your business, less generic.
This is what most "content strategy" courses never deliver: a repeatable feedback mechanism that improves over time without manual analysis.
Run the loop now.
Pull your data. Open your Claude Project. Paste the Content Performance Loop Prompt. Paste your data where the prompt asks. Run it.
Save the output. Take the three recommendations. Use them to plan next month's posts. That is the loop closed. From this point forward, every monthly review runs through this prompt instead of relying on your manual gut read.
Six questions in Lesson 8.3. AI doing the heavy reading here. Either way, monthly.
Module 8 closes here. The next module is where we map your visibility geographically using heat maps to find the neighborhoods where your work is paying off — and the ones where it still has not.
The vocabulary that follows you.
- Content Performance Loop Prompt
- The prompt that takes your GBP data, social analytics, and published post list, then returns what worked, what to stop, and three specific recommendations for next month. Closes the feedback loop most content systems leave open.
- Performance Loop Output
- The five-section AI output (topics, post types, what to do more of, what to stop, three recommendations) saved on its own page in your workbook for use in next month's planning.
- Post types (content categories)
- Education, social proof, conversion, and human / BTS. The four categories the loop sorts your posts into when measuring what performed.
- Feedback loop
- The repeatable mechanism where last month's performance data shapes next month's content decisions. The piece that turns content from a guessing game into a learning system.