You paid for a guest post, the publisher sent a URL, and the link is there. Done? Not quite. A link can be visible in the browser and still be worthless to search engines. Here is the checklist we use when verifying placements, plus a small Node.js script that automates most of it.

1. The page returns 200, not a redirect chain

Request the article URL and look at the final status code. A 301 to the homepage or a 404 means the post was removed or moved. Follow redirects, but record them: one redirect is fine, a chain that ends somewhere else is not.

2. The link is in the raw HTML, not only after JavaScript

Some sites inject or rewrite links client-side. Google renders JavaScript, but many other crawlers (and AI search crawlers) do not. If the link only exists after rendering, treat it as a weaker link. Check the HTML you get from a plain HTTP request first.

3. The rel attribute is what you agreed on

Look at the anchor tag itself:

  • rel="nofollow", rel="sponsored" or rel="ugc" change how the link is treated.
  • None of these is "bad" by itself, but it should match what was promised.

4. The page is indexable

A perfect link on a page that search engines are told to ignore does little. Check:

  • <meta name="robots" content="noindex"> in the HTML
  • an X-Robots-Tag: noindex response header
  • robots.txt blocking the path

5. The canonical points to the page itself

If <link rel="canonical"> points to another URL, search engines may consolidate signals to that URL, and your link may never be counted. Self-referencing or missing canonical is what you want.

6. The anchor text and target URL are correct

Typos happen: a missing https://, a trailing UTM parameter you did not ask for, or a different anchor. Compare the href and the visible text with your order.

7. Re-check later, not just once

Links disappear quietly: site redesigns, expired content deals, plugins that add nofollow to every outbound link. A check on day one proves very little about month six. Schedule the same check weekly or monthly.

A small script that does checks 1 to 6

Node 18+ (built-in fetch) and cheerio for HTML parsing:

npm install cheerio 
// check-link.mjs import * as cheerio from "cheerio"; const [pageUrl, targetUrl] = process.argv.slice(2); if (!pageUrl || !targetUrl) { console.log("Usage: node check-link.mjs <page-url> <target-url>"); process.exit(1); } const normalize = (u) => u.replace(/\/+$/, "").replace(/^https?:\/\/(www\.)?/, ""); const res = await fetch(pageUrl, { redirect: "follow" }); const html = await res.text(); const $ = cheerio.load(html); const report = { status: res.status, finalUrl: res.url, redirected: res.redirected, noindexMeta: /noindex/i.test($('meta[name="robots"]').attr("content") || ""), noindexHeader: /noindex/i.test(res.headers.get("x-robots-tag") || ""), canonical: $('link[rel="canonical"]').attr("href") || null, links: [], }; $("a[href]").each((_, el) => { const href = $(el).attr("href"); let absolute; try { absolute = new URL(href, res.url).href; } catch { return; } if (normalize(absolute).startsWith(normalize(targetUrl))) { report.links.push({ href: absolute, anchor: $(el).text().trim(), rel: $(el).attr("rel") || "(none, followed)", }); } }); report.canonicalOk = !report.canonical || normalize(new URL(report.canonical, res.url).href) === normalize(res.url); report.found = report.links.length > 0; console.log(JSON.stringify(report, null, 2)); 

Run it:

node check-link.mjs https://example-blog.com/some-guest-post https://yoursite.com 

You get one JSON object: status, final URL after redirects, noindex flags, whether the canonical is self-referencing, and every matching link with its anchor text and rel value. Put it in a cron job or a GitHub Action with a list of URLs and you have a basic link monitor.

What the script does not cover

  • robots.txt rules (use a parser such as robots-parser)
  • JavaScript-rendered pages (use Playwright if you need the rendered DOM)
  • Whether the page is actually indexed (check with site: or Search Console)

Why this matters

Most link problems are not dramatic. The post is still there, but a noindex was added, or the link turned into nofollow after a CMS update. A boring, repeatable check catches these before they cost you rankings.

We run this kind of verification on every placement at PublisherBay, a guest post marketplace with 5,000+ vetted publishers, because buyers should not have to check links by hand. If you buy or sell guest posts, feel free to take the script above and adapt it.

What else do you check before you consider a link "live"? Let me know in the comments.