website-health-audit

Check a list of small-business websites in one go and rank them from worst to best.

For every URL it looks at what a customer (and Google) runs into:

• Gone: the domain no longer resolves, the server does not answer, the SSL certificate is broken.
• Expired or parked: registrar "lander" pages (including the JavaScript redirect GoDaddy uses), "Website expired" pages from site builders, domains for sale, "coming soon" placeholders.
• Hijacked: the old domain now shows gambling or pharma spam.
• Outdated: no HTTPS, not mobile-friendly, copyright year 3+ years old, frames/Flash, old jQuery.
• Losing customers: a phone number shown but not tap-to-call, no contact form or online booking, slow first load, no page title or meta description.

Each problem adds points, so the report is sorted by how badly the site needs work. I use it to find local businesses whose website is broken before offering to fix it.

• Standard library only. No requests, no browser. Python 3.10+.
• Reads a text file (one URL per line) or a CSV with a url/website column, de-duplicates and skips blanks.
• Parallel (8 workers by default) with a per-site timeout, and one bad site never stops the batch.

Measured result

$ python site_audit.py sample_urls.txt --md sample_report.md --workers 4
 10  http://this-business-closed-years-ago.invalid  domain does not resolve: the website is gone
  9  http://example.com                             no HTTPS: browsers label it 'Not secure'; no meta description (hurts Google results); no contact form or online booking
  1  https://www.python.org                         old jQuery 1.8
  0  https://demarkstudio.ca

$ python -m pytest -q
............
12 passed in 0.85s

The tests start a local HTTP server that serves a healthy page, an outdated one, a GoDaddy-style JavaScript parking redirect, a spam takeover and a 404, so they run offline in under a second.

On a real list of 187 business websites in one Canadian city it found 7 expired/parked domains, 2 hijacked by gambling spam and 28 that were down (domain gone or server not answering).

Usage

python site_audit.py urls.txt                                  # table in the terminal
python site_audit.py leads.csv --column website --csv report.csv --md report.md
python site_audit.py urls.txt --min-score 6 --timeout 10       # only sites that clearly need work

report.csv columns: score, url, final_url, status, http, seconds, platform, copyright, problems.

Scores are a triage aid, not a verdict: some healthy sites block automated requests (HTTP 403), so open the top results in a browser before contacting anyone.

Author

Marvin Palencia, founder of DeMark Studio, Miramichi, New Brunswick, Canada. I build websites, Stripe payment integrations and Python automations for small businesses. Portfolio: marvin.demarkstudio.ca

MIT License.