Most SEOs track impressions and clicks. Nobody systematically tracks the rate of change of position over time per query. GSC gives you 16 months of data. This strategy uses that data to identify queries in active ranking decline before they fully collapse, and triages which pages need intervention and in what priority order.
Why This Is Non-Obvious
Position drops are usually caught late, after traffic has already cratered. By measuring velocity (not just position), you can intercept a page declining at -0.3 positions/week and intervene 8 to 12 weeks before a competitor overtakes you. Most SEOs look at snapshots. You are looking at derivatives.
Calculate Velocity Score
For each query, calculate a linear slope across the four position data points. Negative slope equals ranking decline.
| Velocity Score | Action Required |
| 0 to -0.1 pos/week | Monitor only |
| -0.1 to -0.3 pos/week | Content refresh within 60 days |
| -0.3 to -0.6 pos/week | Full content rewrite + internal link audit |
| -0.6 or steeper | Emergency: full page overhaul |
Cross-Reference Impression Volume
Prioritize queries where velocity decay is severe AND impressions exceed 1,000 per month. A page declining fast on a low-volume query is not worth the same intervention as one declining on a 50,000 impression query.
Use this prompt after you have exported your GSC data into a CSV or pasted query data directly into Claude.
Prompt
You are an expert SEO analyst. I am going to give you a dataset of keywords from Google Search Console.
Each row contains: Keyword | Position_Now | Position_30d_Ago | Position_90d_Ago | Position_180d_Ago | Monthly_Impressions
Your job is to:
1. Calculate a velocity score for each keyword by computing the linear slope of position change across all 4 time points. A positive slope = ranking is rising. A negative slope = ranking is falling.
2. Classify each keyword into one of these tiers:
– CRITICAL (slope worse than -0.6/week): needs emergency rewrite
– HIGH RISK (slope -0.3 to -0.6): needs rewrite within 30 days
– WATCH (-0.1 to -0.3): needs refresh within 60 days
– STABLE (0 to -0.1): no action
– IMPROVING (positive slope): flag as wins
3. For each CRITICAL and HIGH RISK keyword, give a 2-sentence hypothesis for why it is declining based on the position curve shape. A sharp sudden drop = likely algorithm or content issue. A gradual slow decline = likely competitor gaining ground or new SERP feature.
4. Output a prioritized action table sorted by: CRITICAL first, then by monthly impressions descending within each tier.
Here is my data:
[PASTE YOUR GSC CSV DATA HERE – keyword, positions across 4 dates, impressions]
