Traffic drops get blamed on Google updates the way weather gets blamed on the moon. Sometimes it really is an algorithm change. Far more often, when we audit a site whose traffic slid, the cause is mechanical: something on the site itself is telling Google not to bother. In one audit this summer we found 194 duplicate pages without canonical tags, 110 pages carrying multiple title tags, 196 pages with no meta description, and 150 internal links pointing at dead pages. All on one site. None of it visible to the owner, because a website with technical rot looks perfectly normal in a browser.
Here are the six problems we find most often, ranked by how much traffic each one costs. The ranking matters because fixing in the wrong order wastes months.
1. Indexing failures: the pages Google never sees
Nothing else on this list matters if Google is not indexing your pages. This is the biggest lever because its cost is total: a page that is not in the index gets exactly zero search visits, forever, no matter how good it is.
The usual suspects are a sitemap that was submitted once years ago and never verified, a stray noindex tag left over from a site redesign, or a robots.txt file blocking a directory it should not. We have seen sites publish content faithfully for a year while a stale sitemap kept Google from ever crediting a word of it.
The check: open Google Search Console, go to the Page Indexing report, and read the count of pages that are not indexed and why. Takes five minutes. If the "not indexed" number rivals the "indexed" one, you have found your traffic problem and can stop reading.
2. Duplicate pages and title tag conflicts: Google cannot tell your pages apart
When two or more URLs serve the same content without a canonical tag declaring which one counts, Google splits credit between them or picks one arbitrarily. When a page carries two different title tags, a common defect in themes and page builders that each inject their own, Google chooses one on its own, and it is often the wrong one.
This is the classic templating defect: it ships silently with a theme change or plugin conflict and replicates across every page built from the template. That is how a site ends up with a hundred affected pages overnight while looking untouched. The cost is diluted rankings across the whole site: everything ranks a little worse.
The check: run any crawler (Screaming Frog's free tier covers 500 pages) and sort by duplicate titles and missing canonicals. Numbers above single digits mean a template problem, not a page problem.
3. Internal 404s: dead ends that drain authority
Every internal link that lands on a deleted page throws away two things: the visitor who clicked it and the ranking authority the link should have passed along. A handful is normal housekeeping. Dozens or hundreds usually trace to a single structural cause, a renamed service page, a deleted category, a footer template pointing somewhere that no longer exists, multiplied across every page on the site.
The 150 internal 404s in the audit above traced back to one broken link in a footer template. One fix, 150 errors gone. That is typical: the count looks scary, the cause is usually singular.
The check: Search Console's Page Indexing report lists 404s Google has found. A crawler shows which of your own pages link to them.
4. Missing meta descriptions: winning the ranking, losing the click
Meta descriptions do not directly move rankings, which is why they rank fourth here and not higher. What they move is the click. When a page has no description, Google assembles a snippet from whatever page text it grabs, and the result reads like a ransom note. Searchers skim it and click the competitor whose snippet reads like it was written on purpose.
A page ranking sixth with a written, specific snippet regularly out-earns a page ranking fourth with an auto-generated one. If you rank at all, this is the cheapest click-through gain available.
The check: your crawler from step 2 already produced the list. Write descriptions for money pages first: services, contact, top posts.
5. Page speed: the tax on every visit
Slow pages cost traffic twice. Google's ranking systems use page experience signals, so speed drags positions down a notch in competitive results. Then the visitors you do earn leave before the page paints: bounce rates climb sharply as mobile load times pass three seconds, and most Arkansas small-business sites we test are slower than that on a phone, usually because of oversized images and plugin bloat.
Speed is fifth, not first, because the fix effort is high relative to the gain unless the site is truly slow. Fix indexing and duplication first; they are cheaper wins.
The check: run PageSpeed Insights on your homepage and your top service page. Look at the mobile score, not desktop. Under 50 means speed belongs on your fix list.
6. Orphan pages: content with no doors
An orphan page has no internal links pointing to it. Google finds pages by following links, so an orphan survives only through the sitemap, ranks weakly without internal votes of confidence, and never receives visitors browsing the site. Service pages built years ago and never added to any menu are the usual casualties. The fix costs nothing: link to the page from your navigation, a related post, or the relevant service hub.
The check: crawlers flag orphans by comparing the sitemap against the link graph.
The pattern behind all six
None of these problems announce themselves. The site looks fine, the traffic chart just leans slowly down and to the right. That slow leak is the tell: algorithm updates hit in steps, technical rot drains in slopes.
Stone Path Consulting runs this exact six-point inspection at the start of every SEO engagement, because pouring content and ad spend onto a technically broken site is buying water for a leaking bucket. If your traffic chart has been leaning the wrong way and nobody can tell you why, call 501.232.1017 or email info@stonepathconsulting.com. We will tell you which of the six it is, with the reports to prove it.