Blog
Shorter takes, still no filler.
The guides are the deep write-ups from real fixes. This is the shorter half: commentary on a Google change, a stance we hold that the industry disagrees with, or a best practice that quietly stopped being one. Opinionated on purpose, and still not general SEO advice for its own sake.
Latest
Recent posts.
Common SEO myths, debunked
The hub for the whole series. Every myth we have taken apart, grouped by where it does most damage: ranking factors that are not, penalties that are not penalties, and the convenient numbers mistaken for causes.
Read it Technical SEOTechnical SEO in 2026, in the order that matters
Every guide to this arrives as a flat list, which is the one shape the work cannot take. Four questions in strict sequence, where failing an earlier one makes every later one irrelevant.
Read it Custom toolsFive custom tools worth building, and why most are not
Not a roundup of software to subscribe to. Five patterns of tool worth building on your own site, what each earns, what each costs, and the failure mode nobody includes.
Read it WordPressThe plugin is not why you are not ranking
Yoast, Rank Math, SEOPress and Slim SEO all output the same handful of tags, and that is the whole job. The differences between them are real and none of them are differences in SEO.
Read it Core Web VitalsImproving Core Web Vitals without pretending it is a lever
Three metrics broken into the parts you can actually measure, the short list of fixes that move them, and an honest answer on when to stop.
Read it Core Web VitalsMobile speed on WordPress is four problems, not forty
Almost every slow WordPress site is slow for the same four reasons, in the same order, and the first accounts for most of it. Why stacking optimisation plugins reliably makes it worse.
Read it Structured dataThe structured data types that actually change your listing
Hundreds of schema.org types exist and about seven change how your result looks. Which earn a visible listing, which do invisible entity work worth doing, and which validate perfectly while earning nothing.
Read it AI searchAI Overviews are not the traffic apocalypse you were sold
They change which queries send clicks, not whether search sends clicks. The losses concentrate in short factual queries you were never converting. Which of your pages are genuinely exposed, and the calm response.
Read it Structured dataSchema markup is not a ranking factor
Google has said so outright. What it buys is eligibility for a different-looking result and an entity a machine can resolve. Both are worth having. Neither is a position, and valid markup earns nothing surprisingly often.
Read it Structured dataWhy we stopped adding FAQPage schema
Since Google restricted FAQ rich results to government and health sites in 2023, the markup earns most sites nothing. Why we removed it rather than maintain code for a benefit that no longer exists.
Read it AuditsMost SEO audits are a crawl export with a logo on it
You can generate a 200-page audit in ninety seconds. The work of an audit is the opposite of listing everything: it is deciding what matters, and ranking the handful of problems actually costing you traffic.
Read it Content strategyGoogle does not penalise AI content, it penalises content with nothing in it
There is no authorship penalty and no detector accurate enough to build one on. What gets acted on is publishing at scale with nothing to offer, which was already true when the same pages were written by people.
Read it MetricsDomain age is not a ranking factor
Old domains do rank better, and age is not why. It stands in for what a domain accumulates while it sits there. A domain registered years ago and parked is exactly as old and worth nothing.
Read it IndexationAn XML sitemap does not get your pages indexed
It tells Google which URLs exist. Whether to crawl, index or rank them are three separate decisions it has no say in. Why resubmitting changes nothing, and how a dishonest lastmod costs you the trust it carries.
Read it Indexation404 errors do not hurt your rankings
A 404 is the correct answer for something that is gone, and Google expects them. The count is never the signal. The four cases that genuinely matter, and why redirecting them all to the homepage is worse.
Read it On-pageWord count is not a ranking factor
There is no minimum and no ideal length. Long pages do rank well, and length is not why: it is the by-product of covering a subject properly. Padding to hit a target puts words between the reader and the answer.
Read it StrategyTraffic is not the goal, and more of the wrong traffic hurts
Traffic is a means, not an end, and the easiest number to grow in the ways that do not help. The wrong visitors cost money, muddy your data and crowd out the buyers. Aim at the outcome and let traffic be one input to it.
Read it MigrationA 301 does not lose 15% of your link equity
Google confirmed no PageRank is lost through a 30x redirect. The 15% is a misremembered damping factor. The risk was never the redirect, it is where you point it and how many hops it takes.
Read it Content strategyYou do not need a blog post for every keyword
A keyword is a phrasing; an intent is what the person wants. Build a page per phrasing and you get thin pages competing with each other for the same result. One strong page beats five weak ones, and consolidation usually beats production.
Read it On-pageLSI keywords are not a thing
A 1980s technique for small document collections that Google has said outright it does not use. The instinct underneath is sound, which is exactly why a wrong explanation has survived this long.
Read it Core Web VitalsCore Web Vitals are a tiebreaker, not a ranking lever
A real ranking signal and a small one. It decides close calls between comparable pages, not whether a weak page ranks. Where speed work pays back is the visitor, not the position, and that is a better reason anyway.
Read it IndexationPage with redirect is not an error, it is a description
It means the URL redirects, so Google indexed the destination instead. That is what a redirect is for. It only worries people because it sits under "not indexed". The three cases where it is genuinely worth a look.
Read it IndexationRobots.txt does not stop indexing, and blocking often makes it worse
Blocking a page in robots.txt removes the ability to read it, not the ability to list it. And adding a noindex to a blocked page cannot work, because Google has to read the page to see the instruction telling it not to.
Read it IndexationYou do not need to submit your site to Google
Google finds sites by following links, so it discovered you the moment one page it crawls linked to you. The submit-to-500-engines services sell a solved problem. What the real tools do, and what actually gets a page indexed.
Read it IndexationThe Google sandbox is a myth, but new sites really do wait
There is no penalty box on a timer. What people call the sandbox is a new domain with none of the signals Google ranks on yet. That reframe is the difference between waiting for a calendar and going to earn them.
Read it MetricsDomain Rating is a scoreboard, not a strategy
DR is an Ahrefs metric, not a Google ranking factor. A useful proxy and a terrible target, because the moment you aim at it the cheapest way to move the number is to buy the links it rewards.
Read it Link buildingThe disavow file is a loaded gun, and most people shoot their own rankings
Google already ignores most spam links, so routine quarterly disavowing mostly neutralises links that were already neutral. Run off a toxicity score, it can demote the earned links that were helping. The two cases where it is genuinely right.
Read it Link buildingNofollow links still have value
Nofollow is a hint now, not a firewall, and a link was never only a pipe for authority. It sends real people, builds brand and makes a profile look earned. Chase the coverage, not the attribute.
Read it PenaltiesYou are not being penalised, you are being outranked
A real penalty is rare and announces itself in Search Console. Most drops are a competitor improving, an update rebalancing, or something breaking on your own site. The label you pick decides the whole investigation.
Read it IndexationDuplicate content is not a penalty, it is a choice Google makes for you
There is no duplicate content penalty. Faced with two equivalent pages, Google picks one to show and sets the other aside. The real cost is it may pick the wrong URL, or spend crawl budget on the copies. Make the choice for it.
Read it AnalyticsBounce rate is not a Google ranking factor
It is a number your analytics invents, Google cannot see it, and even as a diagnostic it means five different things you cannot tell apart. What to measure instead, and the dwell-time half-truth.
Read it On-pageMeta descriptions do not affect rankings, and Google rewrites most of them
Not a ranking factor, and Google rewrites the majority anyway, generating a snippet to fit the query. Write good ones for the few pages whose clicks you are actually fighting for, and let the long tail go.
Read it On-pageKeyword density is not a thing
There is no magic percentage and there never really was. Google does not count how often a word appears, the tools that grade your density invented the target, and writing to a ratio makes the page worse.
Read it Content strategyPublishing frequency is not a ranking factor
There is no bonus for posting twice a week. Freshness belongs to the query, not the publisher. A cadence target reliably produces the thin, overlapping pages that hold a site back.
Read it IndexationCrawl budget is probably not your problem
Google has said most sites never need to think about it. It is a satisfying diagnosis because it is technical and external, which is exactly why it hides the real cause.
Read it On-pageYou can have more than one H1
Google has said plainly that multiple H1s are fine. The rule survives because counting elements is easy. The heading faults that actually cost you are the ones no counter catches.
Read it Link buildingSocial signals are not a ranking factor
Shares are not counted and the links are nofollow almost without exception. Social still matters for search, through distribution and brand demand rather than the mechanism being sold.
Read it MetricsYour ranking is not a single number
Results vary by location, device and history, so there is no one position to report. What a tracker actually measures, why Search Console disagrees with it, and the figure worth reporting instead.
Read it On-pageGoogle rewrites your title tags too
Everyone knows it rewrites descriptions. It has been generating result titles since 2021. What triggers a rewrite, why it is not a rejection of your tag, and where the effort belongs.
Read it AccessibilityAlt text is not a keyword slot
It exists so someone who cannot see the image knows what it shows. Stuffing it is named as spam in Google's own guidelines, and decorative images should be empty, not filled.
Read it StrategyExact-match domains are not a shortcut
It worked, which is why the idea survives, and Google targeted it in 2012. Keywords in a URL are a very small factor, and the branding cost you pay for them is permanent.
Read itGuides vs blog
Which is which.
The guides are long, evergreen, and pulled from a specific problem solved on a live site. They are the ones worth bookmarking.
The blog is shorter and more opinionated: a reaction to a Google change, a position we hold, or a piece of received wisdom worth arguing with. It moves faster and it is happy to take a side.
Neither is a place for general SEO advice written to catch traffic. If a post does not have a point we actually believe, it does not go up.
Keep reading
Elsewhere
Book me
Disagree with one of these?
Good. If you think a stance here is wrong, tell us why. We would rather be corrected than comfortable, and the best posts come out of exactly that.