Programmatic Set Analyzer
Groups your Search Console pages by URL shape and prints the distribution inside each set: how many pages carry the clicks, how many earn nothing, and which of the dead ones still get impressions. The number every programmatic case study leaves out.
- Runs entirely in your browser, nothing uploaded
- Separates no-demand pages from title problems
- Gini and Pareto per template, so sets are comparable
How to use it
- In Search Console, open Performance → Pages, set the widest date range available, and click Export → Download CSV. You want Pages.csv from the zip.
- Drop it here. Pages are grouped by URL shape, so /tools/merge-pdf and /tools/split-pdf collapse into one template.
- Read the dead share per template, then open a template to see which zero-click pages still collect impressions. Those are the only ones worth rewriting.
What the numbers mean
- Dead share
- The percentage of pages in a set that are not winners. High is normal, not broken: in the reference dataset 6 of 85 generated pages took 82.9% of the clicks and 22 earned nothing at all. The mistake is not that most pages fail, it is expanding the set before reading which ones did.
- Zero clicks with impressions
- The page is being shown and passed over. Demand exists and the result is not winning it, which is a title and snippet problem and fixable in an afternoon. Zero clicks with zero impressions is the other thing entirely: nobody searched, and no rewrite will help.
- Gini
- How unequal the click distribution is, from 0 (every page equal) to 1 (one page takes everything). Useful because “6 of 85” is vivid but not comparable between sites with different page counts.
- Pages to 80%
- How many of your top pages it takes to reach 80% of the set’s clicks. If the answer is a handful out of hundreds, you have found your winners and the rest is maintenance.
Questions
Is a high dead share a failure?
No. It is the ordinary shape of a programmatic set, and every case study that omits the distribution is describing its winners and quietly not mentioning the rest. What the number should change is your next decision: read the distribution before each expansion rather than after.
What should I do with the dead pages?
Split them. The ones with impressions have confirmed demand and a snippet that is losing, so rewrite the title. The ones with no impressions were never searched for, and the honest options are removing them or accepting they are permanent overhead on your crawl budget.
What counts as a template?
Pages sharing a URL shape once the varying segment is starred, so /guides/a and /guides/b become /guides/*. Numeric, hash-like and date-like segments are always treated as variable. You can group by the last segment or the last two.
Is my data uploaded?
No. The export is parsed and analysed in your browser. There is no server, no account and no storage.
Guides that use this tool
- Pages are indexed but get no traffic
- Programmatic SEO: what it is, and the distribution nobody prints
- Programmatic SEO mistakes that get whole sites handled as a block
- Product-led SEO: when the page is the product
- Should your SaaS have a blog? Only under one condition
- Should you delete old content? Usually update, sometimes remove
Other free tools
- Search Console Analyzer: Drop your export, get a verdict
- SERP Audit: Paste a URL, get the fix list in priority order
- Schema Generator: Valid JSON-LD, with the guard rails on
- SERP Preview: Pixel-accurate, not character-counted
- Position Value Calculator: What is rank 4 actually worth?
- Schema Inventory: Which structured data is on which page
- Internal Link Graph: Which pages nothing points at
- OG Image Generator: And what it looks like once each platform crops it
- Noise or Signal: Did that actually do anything?
- Crawler View: What lands in the HTML, before anything runs
- Log File Analyzer: What Googlebot actually crawled