A Google algorithm update is a change to how pages get ranked, and a few times a year one is large enough to be named. Traffic moves, sometimes sharply, and the temptation is to change everything at once.
That is almost always the wrong response. This article gives you a calmer process.
First, check it really is a google algorithm update
Traffic drops for many reasons and most of them are not an algorithm update. Before assuming anything, rule out the ordinary explanations.
Did somebody publish a change to the site that week? Perhaps a page got accidentally blocked or set to no index. Tracking breaks more often than people expect, in which case the traffic is fine and the reporting is not. Seasonality explains a surprising number of drops, and a year on year comparison settles that in seconds.
Only once those are excluded is an update the likely cause, and the date will usually line up with one that has been announced.
Work out what actually moved
A sitewide drop and a drop on one section are completely different problems, so measure before theorising.
Open Search Console, compare the weeks either side, and group the losses. Did every page fall a little, or did a handful fall a lot? Are the losses concentrated in one part of the site, or one kind of page, or one kind of search?
Very often an algorithm update turns out to have affected a narrow slice, and the sitewide number was hiding that. Fixing the wrong section is worse than doing nothing.
Wait before you change anything
This is the hardest advice to follow and the most valuable. A large update takes days or weeks to finish rolling out, and rankings move about while it does.
Consequently a change made in the first week is made against numbers that were still settling. Worse, if the traffic recovers on its own, you will credit the change and repeat it next time.
Two weeks of patience costs you very little. Rewriting forty pages in a panic can cost a great deal.
What a google algorithm update usually rewards
The named updates of recent years point consistently in one direction, whatever the details.
Each algorithm update in that run has favoured the same things. They favour pages written for people rather than for search engines, pages with genuine experience behind them, and sites that are clearly what they claim to be. Nothing there is new, since the basics that do not change have pointed the same way for years.
They tend to punish the opposite: thin pages produced at volume, content assembled from other search results, and sites covering subjects they have no business covering.
So the honest question after an update is uncomfortable. Would a knowledgeable person consider this page the best answer available? If not, that is the work, and no technical adjustment substitutes for it.
How recovery from an algorithm update works
Recovery is rarely a switch. Sites that come back usually do so at the next broad algorithm update rather than the moment they fix something, because that is when a search engine recalculates its assessment.
That means the loop is slow: improve genuinely, wait, see what the next update makes of it. Anybody promising recovery on a timetable is describing a hope.
What not to do after a google algorithm update
- Never delete pages in bulk because a tool called them thin. Read them first.
- Leave your site structure alone while rankings are still moving.
- Buying links to compensate adds a second problem to the first.
- Rewriting everything at once teaches you nothing from the result.
- A forum thread describing a different site in a different market is not evidence about yours.
Tell the difference between a drop and a shuffle
Rankings move every day without anybody noticing, and a page bouncing between positions four and seven is normal weather rather than a signal. A drop worth investigating has three features: it starts on a particular day, it holds for more than a week, and it affects a group of pages rather than one.
Checking clicks rather than average position helps here. Average position blends every search a page appears for, so a page can look worse while its actual traffic is unchanged. Clicks are the number that pays.
Where a google algorithm update really has hit, the shape is usually obvious once you group the losses: a whole section, or a whole kind of page, moving together on the same date.
Building a site that minds updates less
The sites that ride out an algorithm update comfortably tend to share a few traits. They cover a defined subject rather than everything. Their pages are ones a person would actually recommend. Links arrive because somebody wanted to cite them. No single page carries most of their traffic.
None of that is a trick, and all of it takes time. That is precisely why it holds up when the rules shift.
Frequently asked questions
How do I know if a Google algorithm update hit my site?
Rule out the ordinary causes first: a site change published that week, a page accidentally blocked, broken tracking, or normal seasonality. If none of those explain it and the date lines up with an announced update, an update is the likely cause.
How long does an algorithm update take to finish?
A broad update usually takes days to a few weeks to roll out fully, and rankings move around while it does. Any change you make during that window is made against numbers that have not settled, which makes it impossible to learn anything from the result.
Should I change my site straight after an update?
No. Wait until the rollout finishes, then measure what actually moved before deciding anything. Very often a sitewide drop turns out to be concentrated in one section, and fixing the wrong part of a site is worse than doing nothing at all.
What do recent algorithm updates reward?
Consistently, pages written for people rather than for search engines, with genuine experience behind them, on sites that clearly are what they claim to be. They tend to punish thin pages produced at volume and content assembled from other search results.
How long does recovery take?
Usually until the next broad update, because that is when the assessment gets recalculated. Sites rarely recover the moment they fix something. Anybody offering recovery on a timetable is describing a hope rather than a process.
Should I delete thin pages after an update?
Read them before deciding. Tools flag pages as thin using word counts, which says nothing about whether a page answers its question well. Some short pages are perfect, and deleting them in bulk removes links, traffic and history you cannot get back.
Can I recover by building more links?
Not if the problem is content quality, and buying links to compensate adds a second problem to the first. Links help a page that deserves to rank. They do not rescue one that a search engine has decided is not the best answer available.
How do I make my site less vulnerable to updates?
Cover a defined subject properly rather than everything shallowly, publish pages a person would recommend, earn links because somebody wanted to cite you, and avoid depending on one page for most of your traffic. None of it is quick, which is why it lasts.