In January 2026 we published a list of SEO mistakes on this blog. By October, the list had made one of those mistakes itself. It told readers to improve First Input Delay, a Core Web Vital that Google retired in March 2024. It carried an “Admin” byline. It also linked out to sites that had no place in an SEO guide. If a reader had followed it, they would have spent hours on the wrong problems.
So we rewrote it. The new version doubles as a map of the common SEO mistakes we keep finding in site audits this year. Some are old, like writing for a keyword instead of the person searching it. Others are new, like blaming rankings for clicks that an AI Overview absorbed, or paying for “AEO” tricks that Google says it ignores. Each section below names the mistake, shows how to spot it on your own site, and gives the fix.
The most common SEO mistakes right now are misreading traffic drops caused by AI Overviews, publishing commodity or unedited AI content, chasing AEO hacks like llms.txt, ignoring search intent, letting pages cannibalize each other, optimizing the retired FID metric instead of INP, leaving pages unindexed, buying unqualified links, hiding authors, and treating every Google update as a penalty.
What Are the Most Common SEO Mistakes Right Now?
The most common SEO mistakes in late 2026 fall into three groups. Measurement mistakes make you fix the wrong problem. Content mistakes make pages easy to skip. Technical and link mistakes stop good pages from ranking at all. The table below lists all ten, with the first place to look for each.
| # | Mistake | Warning sign | Where to check | Fix effort |
|---|---|---|---|---|
| 1 | Blaming rankings for an AI Overview click loss | Position steady, CTR falling | Search Console Performance report | Low |
| 2 | Publishing commodity content | Pages indexed but earning no clicks | Search Console Pages and Performance reports | High |
| 3 | Chasing AEO and GEO hacks | Budget spent on llms.txt or “AI schema” | Your task list and invoices | Low |
| 4 | Ignoring search intent | Ranks briefly, then drops | A live search for the keyword | Medium |
| 5 | Keyword cannibalization | Two URLs swapping for one query | Performance report, Pages tab | Medium |
| 6 | Optimizing FID instead of INP | Old checklists, poor INP field data | Core Web Vitals report | Medium |
| 7 | Pages missing from the index | “Crawled, currently not indexed” | Page indexing report | Low |
| 8 | Unqualified paid links | Exact-match anchors on off-topic sites | Backlink tool, anchor report | High |
| 9 | Anonymous authorship | “Admin” or no byline | Your post template | Low |
| 10 | Treating updates as penalties | Panic edits mid-rollout | Search Status Dashboard | Low |
How to use this list
Start with the mistakes marked low effort. Numbers 1, 7, and 9 take an afternoon each and often explain more of a traffic drop than any content rewrite. Then move to intent and cannibalization, because both decide which page Google even considers. Leave the high-effort fixes for last.
One warning before you start. Don’t fix all ten at once. If you change titles, content, internal links, and redirects in the same week, you won’t know which change helped. Work in batches, note the date of each change, and give Google a few weeks to recrawl before you judge the result.
Mistake 1: Blaming Rankings for a Traffic Drop That AI Overviews Caused
A traffic drop isn’t always a ranking drop. If your average position held steady while clicks fell, an AI Overview is often answering the query on the results page. Rewriting a page that never lost its position wastes months, so separate the two causes before you change anything.
The size of this effect is now measured. Ahrefs studied 300,000 keywords and found that an AI Overview correlated with a 58% lower click-through rate for the top-ranking page, using December 2025 data (Ahrefs, February 2026). The same study a year earlier put the drop at 34.5%. Pew Research Center tracked real browsing behavior. Users clicked a traditional result on 8% of visits when an AI summary appeared, against 15% when none did (Pew Research Center, July 2025).
That doesn’t mean SEO stopped working. It means clicks and rankings are now two separate numbers, and you need to read both. You can see the full methodology in the Ahrefs AI Overviews click-through study.
How to tell a ranking problem from an AI Overview problem
Open the Search Console Performance report and compare the last three months with the same period a year earlier. Then sort your top queries into two groups.
- Position dropped and clicks dropped. This is a ranking problem. Keep reading this guide.
- Position held but CTR fell. Search the query yourself. If an AI Overview sits above your result, the lost clicks went there.
Google also launched a Generative AI performance report in Search Console on June 3, 2026. It shows how often your pages appear inside AI Overviews and AI Mode. It reports impressions only, not clicks, so use it to confirm visibility rather than to measure traffic.
What to do about it
Shift effort toward queries where people still click: comparisons, pricing, tools, and anything that needs a decision. Track leads and branded searches next to organic sessions. A page that gets cited in an AI Overview and drives one qualified inquiry is worth more than ten informational visits.
Mistake 2: Publishing Commodity Content That Anyone Could Write
Commodity content repeats what the first few results already say. Google’s own 2026 guidance on generative AI search contrasts it with non-commodity content built on first-hand experience. When a page offers nothing a reader can’t get from a chatbot, it struggles to rank and rarely gets cited.
Google uses a blunt example. A post called “7 Tips for First-Time Homebuyers” is commodity content. A post explaining why one buyer waived an inspection and what they found in the sewer line is not. The difference is experience the writer actually has.
Scale makes this worse. Since the March 2024 spam update, Google’s scaled content abuse policy covers many pages made mainly to manipulate rankings, no matter how they’re created. That wording matters. The policy doesn’t care whether a person, a content farm, or an AI tool produced the pages. Google’s guidance on AI content says generative AI can help with research and structure, but mass-producing pages without adding value may break the policy.
The test we use before publishing
Ask one question: could a reader get this from the top three results or from ChatGPT in thirty seconds? If the answer is yes, the page needs something only you can add. That could be numbers from your own campaigns or a screenshot of a real report. It could be a mistake you made and fixed, or a firm opinion you can defend with a source.
Where AI fits
AI drafts are fine as a starting point. Unedited AI drafts are the problem. Fact-check every claim, cut generic filler, and add the specific detail a practitioner would notice is missing. If you outsource writing, judge writers on their sources and first-hand detail, not word count. That’s the standard we hold our own SEO content writing service to.
Mistake 3: Chasing AEO and GEO Hacks That Google Says It Ignores
Google’s guide to generative AI search says optimizing for AI Overviews and AI Mode is still SEO. It also names tactics you can ignore for Google Search: llms.txt files, chunking content into tiny pieces, rewriting pages for AI, chasing fake mentions, and special AI schema.
This is the contrarian point many “AEO” guides skip, including some of our own. We published several answer engine optimization posts in 2025 that treated schema as an AI lever. Google’s documentation, last updated in July 2026, says structured data isn’t required for its generative AI features and that no special schema.org markup exists for them. Schema still helps you qualify for rich results, so keep it. Just stop treating it as an AI Overview switch. Our comparison of AEO and SEO explains how the two overlap, and we’re updating it to match Google’s guidance.
There’s a fair caveat. Other AI systems, such as ChatGPT or Perplexity, may read files Google ignores. Google says an llms.txt file will neither help nor harm your Google visibility. So if you keep one for other tools, fine. Just don’t pay for it as a Google tactic.
What actually affects AI visibility in Google
According to the Google guide to optimizing for generative AI features, these are the levers that matter:
- Indexing. A page must be indexed and eligible to show a snippet. Your site must also be included in Search generative AI features in Search Console.
- A unique point of view. First-hand experience beats summaries of other pages.
- Clear structure. Headings and short sections help readers first, and AI systems follow.
- Images and video. AI features can pull these in when they support the text.
- Local and product data. Google Business Profile and Merchant Center feed AI answers for local and shopping queries.
Mistake 4: Writing for the Keyword Instead of the Search Intent
Search intent is the job a searcher wants done. If Google shows product pages for a query and you publish a how-to guide, the page might rank for a week and then fade. Check the live results before you write, then match their format, depth, and angle.
Most intent mistakes come from assuming. A writer sees “best CRM for small business” and writes a definition of CRM software. The searcher wanted a short list with prices and a recommendation. The page answers a question nobody asked.
| Intent | What the searcher wants | Clue in the results | Format that fits |
|---|---|---|---|
| Informational | To learn or fix something | Guides, AI Overview, People Also Ask | How-to or explainer |
| Commercial | To compare before buying | “Best” lists, reviews, comparison tables | Comparison or shortlist |
| Transactional | To buy or sign up now | Product pages, prices, shopping ads | Product or service page |
| Navigational | To reach a specific site | One brand dominating the page | Brand or login page |
Our five-minute intent check
Open a private browser window set to the US and search your keyword. Then note five things:
- What type of page holds the top five spots
- Whether an AI Overview appears
- The People Also Ask questions
- How long and detailed the top pages are
- The angle the top three share, and the angle none of them cover
The first four tell you what format Google expects. The fifth is where your page earns its place. This article is an example. The results for “common SEO mistakes” are almost all listicles, so this is a list. Few of them cover AI Overview click loss or Google’s 2026 AI search guidance, so those sections lead.
Mistake 5: Letting Your Own Pages Compete for the Same Query
Keyword cannibalization happens when two or more pages on one site target the same search intent. Google usually picks one to show, often not the one you’d choose, and may rotate between them. Merging them into one stronger page usually beats keeping both.
We found this on our own blog during a 2026 audit. We had several posts targeting local link building and two targeting link building for law firms. Each was decent alone. Together they split links and clicks, and none ranked as well as one combined guide would. Those merges are now on our cleanup list.
Cannibalization usually grows out of good intentions. A team publishes a new post on a topic every few months, each targeting a slightly different phrase. Google sees the same intent behind all of them. The fix is a keyword map: one row per intent, one URL per row, checked before anyone writes a new post.
How to find it in Search Console
Go to the Performance report and filter by one query. Then open the Pages tab. If two or more URLs earn impressions for that query, and their positions swap week to week, you’ve found cannibalization. Repeat for your top twenty queries.
How to fix it
- Pick the stronger page, judged by links, clicks, and how well it matches intent.
- Move the best sections of the weaker page into it.
- Redirect the weaker URL to the stronger one with a 301.
- Update internal links so they point straight to the surviving page.
- Use a canonical tag only for true duplicates you must keep live, such as filtered product pages.
Our full guide to fixing keyword cannibalization walks through the edge cases, including when two similar pages can both stay because their intents differ.
Mistake 6: Optimizing for FID, a Core Web Vital Google Retired
Interaction to Next Paint (INP) replaced First Input Delay (FID) as a Core Web Vital on March 12, 2024. Many checklists, including the earlier version of this article, still tell readers to improve FID. Search Console stopped reporting FID at the switch, so time spent on it fixes nothing.
INP is harder to pass than FID was. FID measured only the delay before the browser started handling a visitor’s first tap or click. INP looks at responsiveness across the whole visit and reports one of the slowest interactions. A page can feel fast on load and still fail INP when someone opens a menu or a filter. The web.dev announcement of INP explains the change in detail.
| Metric | What it measures | Good score |
|---|---|---|
| Largest Contentful Paint (LCP) | Loading speed of the main content | 2.5 seconds or less |
| Interaction to Next Paint (INP) | Responsiveness to clicks, taps, and keys | Under 200 milliseconds |
| Cumulative Layout Shift (CLS) | Visual stability while loading | Under 0.1 |
Thresholds as listed in Google Search Central documentation, last updated December 2025.
Where INP usually breaks on WordPress sites
On the WordPress sites we audit, the usual causes are heavy page builders, chat widgets, cookie banners, and stacks of tracking scripts. Each one adds JavaScript that competes with a visitor’s click. Start by removing scripts nobody uses. Then delay the rest until after the page becomes interactive.
Check field data, not just lab scores. PageSpeed Insights shows real-user INP at the top when your site has enough traffic. Lab tests often miss slow interactions because they don’t click anything. Our guide to measuring and improving Core Web Vitals covers each metric step by step.
Keep speed in proportion, too. Google says good Core Web Vitals align with what its ranking systems reward. Still, a fast page with commodity content loses to a slower page that answers the question better.
Mistake 7: Accidentally Hiding Pages From Google
A page that isn’t indexed can’t rank, and it can’t appear in AI Overviews either. Stray noindex tags, robots.txt blocks, and stale sitemaps hide more pages than penalties ever do. Check indexing for every new post within a week of publishing it.
This happened to us in October 2026. Our three newest posts were live on the blog but missing from our Rank Math post sitemap. Nothing was blocked. The sitemap cache simply hadn’t refreshed. The fix took two minutes once we spotted it, but nobody would have spotted it without checking.
The lesson is that indexing problems are silent. Search Console doesn’t email you when a post never gets crawled, and your rank tracker just shows nothing. A short routine catches most of them before they cost traffic.
A ten-minute indexing check
- URL Inspection. Paste each new URL into Search Console and confirm it says the page is on Google.
- Page indexing report. Read the “Crawled, currently not indexed” and “Discovered, currently not indexed” lists. Thin or duplicate pages often sit here.
- Sitemap versus reality. Compare your sitemap with your live post list. Missing URLs point to a cache or plugin setting.
- Leftover noindex tags. Staging sites often ship with a site-wide noindex that survives launch. Check the robots meta tag on a few live pages.
- robots.txt. Make sure it doesn’t block folders that hold content, CSS, or JavaScript.
If a page stays out of the index after all that, the cause is usually quality, not a technical block. Google crawled it and decided it wasn’t worth keeping. That sends you back to mistake 2. For more methods, see our guide on how to check if a page is indexed on Google.
Mistake 8: Buying Links That Pass Ranking Credit Without Qualifying Them
Google’s spam policies list buying or selling links that pass ranking credit as link spam. That includes paid guest posts with keyword-rich anchors. Google allows paid links when they carry rel=”sponsored” or rel=”nofollow”. Leaving that attribute off is a risk the site owner takes on knowingly.
We sell guest posts and link insertions, so we see both sides of this. Here’s the honest version. Google’s spam policies for web search name paid links, excessive link exchanges, automated link programs, and low-quality directory links as link spam. Most buyers still ask for links without the sponsored attribute, because those are the links that might pass ranking credit. That’s a bet against a published policy, and Google can discount the links or issue a manual action. Anyone who calls that risk-free is selling something.
The risk isn’t the same for every link. It climbs with patterns Google’s systems are built to catch.
Red flags in a link offer
- The same exact-match anchor across many placements
- A host site that publishes anything for anyone, on any topic
- Articles written only to hold a link, with no real reader in mind
- Sites that list “write for us” pages in every niche at once
- Any offer that promises a specific ranking or traffic result
Lower-risk placements share the opposite traits. The host site is relevant to your business, and its editors review and reject submissions. Anchors are branded or natural, and the article is worth reading without the link. Our posts on how Google detects manipulative guest post links and guest posting after the Helpful Content Update go deeper on both.
Publishers carry risk too. Google’s site reputation policy covers third-party content placed on a host site mainly to borrow its ranking signals. A blog that accepts any paid post, on any topic, with no editing, fits that description.
Mistake 9: Publishing Without a Real Author
An “Admin” byline tells readers nothing about who stands behind the advice. Google’s guidance on helpful, people-first content asks who created a page and why. Name a real author, link a short bio, and show the experience behind the claims. It’s one of the fastest trust fixes available.
The earlier version of this article ran under “Admin.” That byline is common on agency and small business blogs. It’s easy to miss because it comes from the theme, not the post. Readers notice it, though. So do the people who rate search quality for Google, and they’re told to look for who’s responsible for a page.
What a useful author signal looks like
- A real name on every post, linked to an author page
- A two or three sentence bio covering role and relevant experience
- A link to a LinkedIn profile or other public work
- Specific first-hand examples inside the article itself
- A visible “last updated” date when the content changes
Don’t fake any of it. Invented experts, AI-generated headshots, and made-up credentials are worse than an honest “Admin” byline, because they read as deception. Author pages need care too. Keep them indexable, list the author’s posts, and use the same name and photo everywhere they publish. That consistency helps readers connect the person to the work.
Our guide to E-E-A-T and why it matters for SEO covers the wider set of trust signals.
Mistake 10: Treating Every Google Update as a Penalty
Core updates reassess quality across the web. They aren’t penalties, and Google says no single fix exists for them. Spam updates are different. They enforce specific policies. Knowing which kind of update coincides with your drop decides whether you improve content or clean up a practice.
2026 has been a busy year. Google ran core updates from March 27 to April 8 and from May 21 to June 2. It also released spam updates in March, June, and August. Another started on September 24 with a rollout window of up to two weeks (Google Search Status Dashboard, 2026). With that many overlapping changes, it’s easy to blame the wrong one.
A calm response plan
- Wait for the rollout to finish. Rankings move around during an update. Judging mid-rollout leads to panic edits.
- Match dates. Compare your drop with the start and end dates on Google’s Search Status Dashboard.
- Check Manual actions. If Search Console shows one, it’s a policy problem, not a core update.
- Group losses by page type. If all your thin listicles fell and your detailed guides held, the update told you what it values.
- Improve, then wait. Recovery from a core update often shows up around a later update, not the next week.
The worst move is rewriting everything at once. You lose the pages that were working and learn nothing about what changed.
Spam updates call for a different response. If your drop lines up with one, read Google’s spam policies with your own site in mind. Look for scaled pages, paid links without attributes, or third-party content you host without editing. Fix the practice first. Google says its systems may take months to confirm a site now complies.
Common SEO Mistakes: Frequently Asked Questions
These are the questions people ask most about common SEO mistakes, drawn from People Also Ask results, Reddit, and Quora.
What are the most common SEO mistakes?
The most common SEO mistakes are ignoring search intent, publishing thin or commodity content, letting pages cannibalize each other, and leaving pages unindexed. Newer ones include misreading AI Overview click loss as a ranking drop and optimizing the retired FID metric. Paying for AEO tactics like llms.txt, which Google says it doesn’t use, is another. Most can be spotted in Search Console in under an hour.
What ruins your SEO the fastest?
The fastest ways to damage SEO are breaking Google’s spam policies, such as buying links that pass ranking credit at scale or mass-producing low-value pages. A site-wide noindex tag left over from staging can also remove a site from search almost overnight. Most other mistakes slow growth gradually rather than causing a sudden collapse, so check policy risks and indexing first.
Why is my website not ranking on Google?
Start with indexing. Use URL Inspection in Search Console to confirm Google has the page. If it’s indexed, compare your page with the current top results for the same query. Most ranking gaps come from a mismatch with search intent, thin content, or a page competing with another of your own. Sometimes the site simply has too few relevant links to compete.
Does AI-generated content hurt SEO?
AI-generated content doesn’t hurt SEO by itself. Google judges the quality of the page, not the tool that drafted it. The problem is unedited AI content published at scale, which can fall under Google’s scaled content abuse policy. Use AI for research and structure, then fact-check every claim and add first-hand detail, data, or examples that a generic draft can’t provide.
Do I need an llms.txt file for SEO?
No, not for Google. Google’s 2026 guide to generative AI search says Google Search doesn’t use llms.txt files and that creating one will neither help nor harm your rankings. Other AI tools may read the file, so keeping one does no damage. Just don’t treat it as a way to appear in AI Overviews or AI Mode. Spend that time on indexing and content quality instead.
Is First Input Delay still a Core Web Vital?
No. Interaction to Next Paint (INP) replaced First Input Delay (FID) as a Core Web Vital on March 12, 2024, and Search Console stopped reporting FID. A good INP score is under 200 milliseconds. If an SEO checklist or audit tool still tells you to fix FID, treat the rest of its advice with caution too, because it hasn’t been updated.
Can bad backlinks hurt my site?
Google’s systems usually ignore spammy links they detect rather than punishing the site. Links can cause real harm when they come from a deliberate scheme, such as paid links at scale with exact-match anchors. That pattern can lead to a manual action. Check the Manual actions report in Search Console. If it’s clear, cleaning up random spam links rarely needs to be your first priority.
How long does it take to recover from SEO mistakes?
Technical fixes like indexing errors can recover within days to a few weeks once Google recrawls the pages. Content improvements usually take several weeks to a few months. Recovery after a core update often appears around a later core update. A manual action lifts only after you fix the problem and Google approves a reconsideration request, which can take weeks.
How do I find SEO mistakes on my own website?
Start with three free reports in Google Search Console: Page indexing, Core Web Vitals, and Performance. Together they show missing pages, slow templates, and queries where clicks are falling. Add a crawler such as Screaming Frog for broken links, redirect chains, and duplicate titles. Then search your top keywords yourself and compare your pages with what ranks above them. Repeat the check every quarter.
What to Fix First
The earlier version of this article taught a metric Google retired, sent readers to sites that had no place in an SEO guide, and didn’t say who wrote it. Those are three of the ten common SEO mistakes covered here, and none of them took long to fix once we looked.
That’s the pattern with most sites we audit. The biggest losses rarely come from one dramatic error. They come from small things nobody checked. A sitemap stops updating. Two posts fight for the same query. A checklist goes stale, or a traffic chart gets read as a ranking problem when an AI Overview took the clicks.
So start small. Open Search Console today and check three things: pages that aren’t indexed, queries where position held but clicks fell, and URLs that compete with each other. Fix what you find, write down the date, and wait for the data before touching anything else. Then work through the rest of the list one batch at a time.
Which of these ten mistakes is hiding on your site right now?
