Website Traffic Dropped: How to Find the Cause

Traffic dropped. It is one of the most common reasons people get in touch, and it almost always sounds the same: “we did nothing and it suddenly fell.” The problem is that there are a dozen possible causes, and diagnosis usually starts from the worst assumption – “we got penalised.”

The order of checks should actually be the reverse: first rule out everything that has nothing to do with search engines at all.

Step zero: did it actually drop

Half of all “drops” are measurement errors. Check these first:

• The comparison window. Four weeks against five, or a holiday week against a normal one.
• Seasonality. Compare against the same month last year, not against last month.
• Filters in the report. An accidentally saved segment produces a drop out of nowhere.
• The tracking code itself. If analytics disappeared from part of the site after an update, traffic “fell” only in the report.

That last one happens more often than you would think. It is easy to check: look for the analytics code in the source of several different page types.

Step one: which traffic dropped

This is the most important question, and it eliminates most hypotheses immediately.

All traffic falling simultaneously is almost never a search problem. It is either the site or the analytics.

Step two: the shape of the graph

The shape of a drop says more about the cause than its depth does.

• A cliff in a single day. Technical. Something broke: pages got blocked from indexing, URLs changed, the server went down, the counter came off. Look at what was done to the site that day.
• A gradual decline over weeks. Competition or ageing content. Nobody broke anything, you are simply being overtaken slowly.
• A step down followed by a plateau. Looks like an algorithm change, or like losing a set of pages from the index.
• A sawtooth trending downward. Often server instability: part of each crawl ends in errors.

Step three: the technical checklist

A quick pass that covers most real causes. Takes half an hour.

1. The site opens. From different networks and countries, not just yours.
2. The robots file does not block indexing. The classic after moving from a staging server.
3. No noindex meta tag on important pages. The second classic, same origin.
4. URLs have not changed. A theme or plugin update can change URL structure without warning.
5. The number of indexed pages. A sharp fall here is the headline symptom.
6. The crawl error report. A spike in server errors means the site was unavailable to crawlers.
7. The security certificate is valid. An expired certificate kills traffic within a day.
8. The sitemap is reachable and current.
9. Speed has not changed abruptly. A new plugin or third-party script can double response time.

Points two and three are the most frequent causes of a sharp cliff. Check them first; it takes two minutes.

Step four: if the technical side is clean

Then the cause is content or competition, and diagnosis works differently.

• Compare queries before and after. If one cluster of queries fell while the rest held, the issue is that specific topic, not the site.
• Look at the results page for those queries. Often the result type itself changed: service pages replaced by articles or by an answer block.
• Check whether you rewrote anything. A rewritten page that used to carry traffic is the most common self-inflicted cause.
• Look at who is now ranking. New entrants with more thorough material are an answer in themselves.

What not to do

• Do not change everything at once. After five simultaneous changes you will never know which one worked.
• Do not rewrite pages on day one. Diagnosis first.
• Do not buy links “to lift it back.” That does not cure a technical error or return pages to the index.
• Do not panic over one week. Swings of ten to fifteen percent are normal.

The order of operations

In short: confirm the drop is real; identify which channel fell; read the shape of the graph; run the technical checklist; and only then think about content and competition.

In our experience most sharp drops come down to one of three things: indexing left blocked after a site update, changed page URLs, and server unavailability during crawls. All three are checkable in half an hour and fixable in a day. The causes that require months of work are considerably rarer than people assume.