What's broken on real websites — a live leaderboard
A live leaderboard of the issues we surface most often, drawn from our own scan database. Every entry below links to the fix guide for that issue. Findings will populate as more audits complete.
Early data — sample size is still small. Treat the rankings as directional, especially toward the bottom of the leaderboard.
What you're looking at
We aggregate every finding from every Seoxpert audit completed in the last 90 days, group them by canonical issue type, and rank by frequency. Each issue links to the matching fix guide in our public coverage catalog — the same guide our customers see when an audit flags the issue on their own site. No paywall.
No scraping.We don't crawl random sites without consent. Every site represented in this data has been voluntarily audited by its owner via our service.
What's broken most often
Top 0 issue types by total occurrences (minimum 3 occurrences to qualify). Click any row to see the fix guide.
How the leaderboard is built
Window. Findings from scans completed in the last 90 days. Older audits are excluded because the web changes quickly and stale findings overweight historic norms.
Dedup key. Findings are grouped by canonical_issue_id, the same identifier our matcher uses to suppress duplicates across re-scans. Issues that the matcher hasn't promoted to a canonical id yet (Tier-3 unmatched findings) are excluded from this leaderboard.
Floor. An issue must appear at least 3 times in the window to qualify for the leaderboard. This keeps long-tail rare findings from polluting the rankings at small N.
Counts.“Times found” counts every occurrence on every page of every site that triggered the issue. “Audits affected” counts distinct completed scans that contained at least one occurrence — re-scans of the same domain each count once. A separate “site owners affected” figure (the count of distinct customers) appears in the hero strip when the data is loaded; it's the honest sites-not-scans denominator.
Refresh cadence.The page regenerates at most once every 24 hours via Next.js incremental static regeneration. The “last refreshed” stamp at the top reflects the most recent regeneration.
Honesty about sample size.Our scan database is the source of truth, but it's our scan database — not a representative sample of the web. Sites that audit themselves with us skew toward technically-aware operators who already think to look. Treat these as “issues that show up even on sites whose owners care enough to audit,” not “global website statistics.”
Cite this report
Free to cite, free to quote. We'd appreciate a link back to this page so readers can verify the data and see the latest refresh.
Your site probably has 3–5 of the issues above
Run a free audit. The findings list you get back will be directly comparable to the leaderboard above.