Your Website Traffic Dropped. Here’s How to Find Out Why, in Order.

You open your analytics on Monday morning and the line is pointing down. Not gently — noticeably. Last week you had 900 visitors and this week you have 500. The immediate reaction is usually one of two things: assume the worst (“Google has penalised me”) or assume nothing (“it’s probably just a slow week”). Both are guesses, and guesses cost you either sleep or customers.

There is a better approach, and it is not complicated. Traffic drops have a fairly short list of causes, and they can be checked in order — starting with the ones that are most common, easiest to check, and most urgent to fix. Work down the list and you will usually have your answer in twenty to thirty minutes.

The first question is not “why did traffic drop” but “did traffic actually drop.” Two very common false alarms account for a surprising share of panicked Monday mornings.

The first is an incomplete period. If you are comparing “this month” to “last month” on the 8th of the month, this month has eight days in it and last month had thirty-one. The comparison is meaningless. Always compare like periods: seven days against the previous seven days, or this Tuesday against last Tuesday.

The second is a broken or missing tracking snippet. If your analytics code stopped loading — because a theme updated, a plugin changed, a developer redeployed the site, or a cookie banner started blocking scripts — your traffic did not drop, your *measurement* did. This is genuinely one of the most common causes of a sudden, dramatic, overnight drop, and it is the single most important thing to rule out first because a site that is not being measured stays not-measured until somebody notices.

If your site was down for six hours on Wednesday, Wednesday’s traffic is gone and it is not coming back. This sounds obvious, but most small business owners have no idea whether their site went down last week — outages tend to happen at 3am, resolve themselves, and never get mentioned by anyone.

This is exactly what uptime monitoring is for: a check that runs every few minutes and tells you, after the fact, precisely when your site was unreachable. If you can line up the dip in your visitor chart with a stretch of downtime, you have your answer and your next action is a conversation with your host, not a marketing project.

Slowness counts too. A site that technically stayed up but took eight seconds to load will quietly lose a large share of its visitors before they ever register as a visit. Page speed measured from real visits — as opposed to a one-off speed test — will show you whether last week was unusually slow.

A total is an average of several different things, and averages hide the story. Break the drop down by referrer and it usually becomes obvious very quickly. Traffic almost never falls evenly across every channel — one source falls off a cliff and drags the total with it.

What you find here tells you what kind of problem you have:

Think back over the last month. Did you redesign anything? Change your page addresses? Move to a new host? Install or remove a plugin? Have a developer “clean things up”? Every one of these can quietly break something that was working.

The most common self-inflicted wound is changed URLs without redirects. If your services page moved from one address to another and nothing tells search engines and old links where it went, every link pointing at the old address now leads nowhere. A site crawler and broken link detection will find these in a couple of minutes — and dead links are one of the few problems on this list that are both easy to find and quick to fix.

This is also the argument for keeping a written record. If you note every meaningful change — “launched the new homepage,” “switched hosts,” “started the newsletter” — directly on your traffic chart using annotations, then six weeks later, when something looks odd, you are reading history instead of trying to remember it.

Plenty of traffic drops are entirely normal and require no action at all. A tax accountant is quiet in July. A garden centre is quiet in December. A B2B service business is quiet between Christmas and New Year, and quiet again for most of August. School holidays, public holidays, and heat waves all move traffic around.

The way to check this is to compare against the same period last year, not against last month. If July was 30% below June last year too, then this July being 30% below June is not a problem — it is your business. Knowing that is worth real money, because it stops you from spending a panicked afternoon fixing something that was never broken.

People jump to this one first and it should almost always be checked last, because it is the hardest to verify and the slowest to fix. If your site was up, nothing changed, no single referrer collapsed, and it is not seasonal, then it is reasonable to look at your standing in search results.

Search rankings move for ordinary reasons: a competitor published something better, Google adjusted how it ranks pages, or a page that was ranking for a lucky term stopped doing so. It is rarely a punishment and almost never personal. The productive response is not to panic-rewrite everything — it is to look at which of your top pages lost traffic and ask what that page was answering, then answer it better.

One newer factor worth knowing about: a growing share of people now research businesses through AI assistants rather than a search box. If some of your traffic quietly moved there, you would see it as a search decline with no obvious cause. AI traffic tracking separates those visits out so you can tell the difference between “fewer people are looking for me” and “people are finding me somewhere else.”

Print this, or keep it somewhere you will find it the next time the line points down:

The reason a traffic drop feels alarming is almost always that it is discovered late — three weeks of decline noticed all at once looks like a catastrophe, while the same decline noticed on day two looks like a task. Ten minutes a week is the entire difference.

It is also why an AI advisor reading your traffic is worth more than another chart. A chart tells you the number went down. A weekly analysis tells you *which* source went down, whether the site had downtime that week, what changed, and what to do about it — in plain English, on a Monday morning, before it becomes a month-long problem.