URL structure for SEO: capitals, underscores, parameters and other traps

1Check this URL and every internal link 30 seconds

Open DevTools (F12), go to the Console and paste:

Copy// Checks this page's URL, and the internal links on it, for the URL problems search engines trip over
const TRACKING = /^(utm_[a-z_]+|gclid|gbraid|wbraid|dclid|fbclid|msclkid|yclid|igshid|mc_cid|mc_eid|_ga|_gl)$/i;
const SESSION = /^(sid|sessionid|session_id|phpsessid|jsessionid|aspsessionid|cfid|cftoken)$/i;
const LOCALE = /^[a-z]{2,3}(-[A-Z][a-z]{3})?-([A-Z]{2}|\d{3})$/; // /en-GB/ style segments aren't capitals problems
const problems = (u, isLink) => {
  const path = u.pathname;
  let decoded = path;
  try { decoded = decodeURIComponent(path); } catch (e) {}
  const letters = path.replace(/%[0-9a-f]{2}/gi, '').split('/').filter(s => !LOCALE.test(s)).join('/');
  const params = [...u.searchParams.keys()];
  const tracking = params.filter(p => TRACKING.test(p));
  return [
    /[A-Z]/.test(letters) && 'capitals',
    /\s/.test(decoded) && 'spaces',
    /\/\/+/.test(path) && 'double slashes',
    /_/.test(path) && 'underscores',
    /[^\x00-\x7F]/.test(decoded) && 'non-ASCII characters',
    (u.origin + path).length > 115 && 'long (' + (u.origin + path).length + ' characters)',
    (params.some(p => SESSION.test(p)) || /;jsessionid=/i.test(u.href)) && 'session ID',
    // Tracking parameters only matter on your own links; on the page's address they just mean a campaign link was used
    isLink && tracking.length && 'tracking: ' + tracking.join(', '),
  ].filter(Boolean).join('; ');
};
console.table({ 'This page': { url: location.href, problems: problems(new URL(location.href), false) || 'none' } });
const site = location.hostname.replace(/^www\./, '');
const links = [...new Set([...document.querySelectorAll('a[href]')]
  .filter(a => typeof a.href === 'string') // skips SVG links, whose href isn't a string
  .map(a => a.href))]
  .map(href => { try { return new URL(href); } catch (e) { return null; } })
  .filter(u => u && /^https?:$/.test(u.protocol) && u.hostname.replace(/^www\./, '') === site);
const flagged = links.map(u => ({ link: u.href.slice(0, 100), problems: problems(u, true) })).filter(r => r.problems);
if (flagged.length) console.table(flagged.slice(0, 50));
console.log(flagged.length + ' of ' + links.length + ' internal links have URL problems');

It checks this page's address, then every link on the page to the same site, for capitals, spaces, repeated slashes, underscores, non-ASCII characters, length, session IDs and tracking parameters.

It uses the same rules as SEOSnip, including skipping /en-GB/ style folders, and also runs the page checks on every internal link (SEOSnip checks those on the page's own address only). Tracking parameters only count on links, not on the address you arrived at. It only sees links in the page as rendered; it makes no requests.

2Test the variants 2 minutes

Try the page's address in capitals, with a doubled slash, and with an extra ?utm_source=test. Each should either redirect to the real address or have a canonical pointing to it. If they all load as separate pages with no canonical, they're duplicates.

3Look at what Google has indexed 5 minutes

In Search Console, the Pages report lists URLs under reasons such as "Duplicate without user-selected canonical" or "Alternate page with proper canonical tag". Scan those lists for capitals, parameters and session IDs.