GEO Monthly Delta Report
Compare two GEO audits of the same site and generate a client progress report showing exactly what moved, by how much, and what is next
Month three of a GEO retainer is when the client asks what they are paying for. Diffing the baseline audit against the current one answers that in numbers: score up 12 points, three crawlers unblocked, 3 of 5 quick wins shipped, here is next month's target.
Who it's for: GEO and SEO agencies on retainer, freelance consultants running monthly check-ins, in-house marketing leads reporting to executives, account managers defending renewals
Example
"Compare the January and March audits for acme.com" → A monthly progress report with an ASCII score bar showing 32 to 44, category and platform deltas with trend arrows, a crawler status changelog, action-item completion percentages, this month's wins, newly discovered issues, and a projected six-month trajectory
CLAUDE.md Template
New here? 3-minute setup guide → | Already set up? Copy the template below.
npx ccplaybooks add geo-compare# GEO Monthly Delta Report Generator
Compare two GEO audits of the same site, baseline against current, and turn the difference into a client-facing progress report. Every point gained on the GEO score is proof of value, which makes this the strongest retention artifact a GEO agency has.
GEO = Generative Engine Optimization: getting a site cited by AI search platforms (Google AI Overviews, ChatGPT, Perplexity, Gemini, Bing Copilot).
## How to invoke
Ask for any of these:
```
compare <domain>Get new playbooks like this one
One email a week with new Claude Code workflows. Free, like everything here.
No spam. Unsubscribe anytime.
What This Does
GEO is Generative Engine Optimization: getting a site cited by AI search platforms like ChatGPT, Perplexity, and Google AI Overviews. Progress there is slow and invisible unless you measure it.
This playbook diffs two GEO audits of the same domain, baseline against current, and writes the difference as a client progress report. Claude parses both files, computes deltas across six score categories, five AI platforms, and 14 crawlers, tracks which action items shipped, and projects the trajectory forward.
It comes from the geo-compare skill in Zubair Trabzada's geo-seo-claude suite, where it exists as the retention tool for monthly client check-ins.
Quick Start
Step 1: Create a Project Folder
mkdir -p ~/Documents/GEOTracking/audits
Step 2: Download the Template
Click Download above, then:
mv ~/Downloads/CLAUDE.md ~/Documents/GEOTracking/
Step 3: Start Working
cd ~/Documents/GEOTracking
claude
Put audit files in audits/ named <domain>-<YYYY-MM-DD>.md, then say "compare acme.com."
How It Picks Files
Give it a domain and it finds the audits itself: matches <domain>-*.md in the audits folder, sorts by date, uses the oldest as baseline and the newest as current. With only one file on hand, it treats that as the baseline and runs a fresh audit for the current side. With none, it stops and asks you to run an audit rather than inventing a starting point.
You can also pass two file paths directly when you want a specific pair.
What the Report Covers
- Executive summary — two or three sentences on what moved and where to aim next month
- Score progress — an ASCII bar chart of baseline versus current versus target
- Category breakdown — deltas across AI citability, brand authority, content and E-E-A-T, technical foundations, structured data, and platform optimization
- Platform readiness — before and after for Google AI Overviews, ChatGPT Web Search, Perplexity, Gemini, and Bing Copilot
- Crawler changelog — which of the 14 AI crawlers changed from blocked to allowed
- Action plan progress — quick wins, medium-term, and strategic items with completion status and percentages
- This month's wins — the tangible results, stated plainly for the client
- New issues discovered — problems the fresh audit surfaced that the baseline missed
- Six-month trajectory — a table of past months filled in and future months projected
Trend Thresholds
Deltas map to symbols so the direction reads at a glance:
| Delta | Symbol | Meaning |
|---|---|---|
| +5 or more | ▲▲ | Strong improvement |
| +1 to +4 | ▲ | Improvement |
| 0 | ── | No change |
| -1 to -4 | ▼ | Slight decline |
| -5 or more | ▼▼ | Significant decline, needs discussion |
Tips
- A declining score often means the newer audit found problems the first one missed. The template frames those as newly discovered opportunities and names which audit surfaced them, which is more honest than hiding the dip.
- Keep audit files in a consistent format. The parser looks for score markers like
GEO Score: XX/100,AI Citability: XX/100, andGPTBot: Allowed/Blocked. - Deltas take time to show up in traffic. The template notes that AI platforms need 4 to 8 weeks to re-index, which sets expectations before the client asks why the score moved and visits did not.
- Run it on a fixed monthly cadence. The six-month trajectory table only pays off once it has three or four filled rows.
Limitations
- This compares one site to itself over time, not against competitors. Competitor benchmarking lives in other skills in the suite.
- When exact scores are missing from an audit file, Claude estimates them from the written findings. Those estimates are labeled, but they make month-to-month deltas softer than they look.
- Both audits should come from the same methodology. Diffing a detailed audit against a quick snapshot produces deltas that measure the method, not the site.
- Dollar and traffic impact figures are conservative projections, not measurements. Pull real numbers from analytics when you have them.