Every SEO tool tells you what's broken. This one hands you the fixed files.
AuditTag crawls your site, finds the on-page problems, writes the missing tags, and packages everything as a zip you deploy yourself.
Free, no account. Takes about thirty seconds.
Fetching your pages…
Three steps, and you stay in control of all of them
Competing tools connect to your CMS and push changes live. We never ask for that access, because the whole point is that you review the work before it ships.
-
Step 1
Run the audit
Paste your URL. We crawl your pages and check them against the rules search engines actually act on. You get your score, every issue we found by type, and the worst pages in full — free.
-
Step 2
Review the pack
We write the missing titles, descriptions and alt text, then build a zip that mirrors your folder structure. A README lists every single change.
-
Step 3
Deploy it yourself
Diff it against your repo, take what you want, drop the rest. Every corrected file is your original with specific tags changed, not a regenerated page.
What we check, and what we do about it
Rule-based detection, so the audit is deterministic and free. A model is used only where something has to be written rather than found.
Title tags
Missing, duplicated across pages, or long enough to be truncated in results.
rewrittenMeta descriptions
Missing, duplicated, or outside the length Google will actually show.
rewrittenHeading structure
Pages with no H1, an empty H1, several competing H1s, or skipped heading levels.
rewrittenImage alt text
Images with no alt attribute, alt text that is just the filename, and decorative images that should be marked as such.
rewrittenCanonical tags
Missing, relative, pointing at http on an https site, or pointing at a page that does not exist.
generatedSchema markup
Organization, LocalBusiness, Article and Product JSON-LD, plus blocks that are present but invalid.
rewrittenOpen Graph and Twitter cards
The tags that decide whether your links render as a preview or a bare URL in Slack, LinkedIn and WhatsApp.
generatedrobots.txt
Missing, blocking the whole site by accident, or not pointing at your sitemap. Your existing rules are preserved exactly.
generatedsitemap.xml
Missing, stale, or leaving out pages we found while crawling.
generatedBrand identity across the web
Whether your structured data declares the profiles you link to, so mentions of you elsewhere count towards your brand instead of nobody's.
generatedWhether a page can be cited
Long articles with no named author and no date. Nothing quoting them — a person, a journalist, an AI answer — has anything to attribute.
flaggedWhat we don't check: page speed, Core Web Vitals and broken links — none of those can be fixed by handing you a file. And we don't build backlinks. Nobody can hand you genuine links in a zip, and the automated version is what Google's spam policies exist to catch. We do the part that is a file: making sure the mentions you earn land on your brand.
Built for sites you deploy from files
Pay once, for one site
Priced by how big the site is, because that is the thing that genuinely takes more work. No subscription, no account.
The free audit gives you your score, every site-wide problem, a full page-by-page breakdown of your three worst pages, and an honest count of everything else by type. The fix pack adds the rest of the page-by-page detail — and, more to the point, the corrected files.
Questions worth asking
Do you need access to my site?
No. We read your pages the same way Google does, over the public web. There is no login, no API key, no plugin with write access, and nothing to revoke afterwards. The fix pack is a zip file — what happens to it is entirely your decision.
Does this help with backlinks and traffic, or only on-page?
We do not build backlinks, and we would be wary of anyone who says they will. Link building is outreach — it needs a person, and the automated version is exactly what Google's spam policies are aimed at. It also isn't something we could hand you in a zip.
What we do is the part of off-site visibility that is a file. Most of what Google and the AI assistants say about a brand comes from sites you don't own, and those mentions only count towards you if your markup says which accounts are yours. We find the profiles you link to, declare them in your structured data, and flag the pages that nothing can cite because they have no named author, no date and nothing original on them. That is the foundation the links land on. Earning them is still your job.
How do I know the changes are safe to deploy?
Because you can read them. Every corrected file is your original file with specific tags changed — nothing reformatted, nothing regenerated. Run a normal diff against your repo and every line we touched shows up. The README lists each change with the before and after value.
What if a fix needs a human decision?
Then we don't make it. Pages with several H1s, canonicals pointing at another domain, a robots.txt blocking the whole site — these are sometimes deliberate. We flag them in a "needs a human" section with what we would do and why, and leave your file alone.
Is the written copy any good?
It is written from your page's own content, in your page's own voice and spelling convention, and checked against the other pages on your site so nothing comes back duplicated. It will not invent claims your page does not make, and it will not put facts into structured data that aren't on the page. Read it before you ship it — that's what the review step is for.
How does the WordPress output work?
You get a single plugin file. It hooks into wp_head to output the corrected tags and schema, and never edits your theme, so a theme update can't undo it. If Yoast or Rank Math is installed, it detects them and defers rather than fighting over the same tags.
How many pages do you crawl?
Up to twenty-five per audit, seeded from your sitemap where you have one. That covers most small sites completely and gives a representative sample of a large one. We honour robots.txt while crawling.