Most blogs that lose traffic are not penalised. They are outcompeted, usually by sites doing the ordinary things more carefully.
These are the twelve problems I find most often when I audit a blog that has stalled. They are ranked roughly by how much damage they do, and every one of them is fixable without starting over.
Quick Info
1. Publishing volume with nothing original in it
The most common story behind a traffic decline in the last two years: a site started publishing three or four generated articles a week, ranked briefly, then lost visibility across the whole domain, including the posts that used to perform.
The mechanism is straightforward. Content that restates what is already available adds nothing to the index, and a site made mostly of it looks like a site with nothing to offer. The old posts get dragged down with the new.
The fix is unpopular: delete or consolidate the thin posts, keep the ones with something real in them, and publish less.
Twenty posts with your evidence in them will outrank two hundred that could have been written by anybody.
2. Burying the answer
The reader asked a question. The article opens with three hundred words about why the question matters. They leave, and the answer engine extracts a competitor's paragraph instead.
Put the answer in the first hundred words. Everything the old structure put first, the context, the credentials, the story, either goes later or goes.
3. Two posts chasing one query
Keyword cannibalisation is the mistake nobody notices, because both posts look fine. In Search Console you see the symptom: two URLs alternating in position for the same query, neither of them ranking as well as one of them would.
4. Abandoning posts after publication
Content decays. Prices change, screenshots age, the advice stops being current, and competitors publish something newer. A post from 2023 that nobody has touched is losing a little ground every month.
Updating twenty existing posts reliably produces more traffic than publishing twenty new ones, and takes less time. It is the least glamorous work in blogging and the highest return.
5. Writing for keywords instead of questions
Keyword-first content selection produces articles that technically match a phrase and answer nothing. You can spot them by their subheadings, which repeat the keyword in four variations and cover the same ground each time.
Start from the question a person would ask out loud. The keyword follows from it; the reverse does not work.
6. No internal linking structure
Blogs accumulate posts without connecting them, so every article is an island and the site never demonstrates depth on anything.
7. Weak or missing About and author information
A site with no named author, no photograph and no explanation of who is behind it struggles in every subject where trust matters, and increasingly in ones where it did not used to.
This is an afternoon of work: a real About page, a byline that links to it, and a short author box on each post explaining why you are qualified to write it.
8. Ignoring the click-through problem
Plenty of blogs rank respectably and get very few clicks, because the title says nothing and the description says less. Impressions are high, click-through is under 2%, and the owner concludes the ranking is not working.
Rewriting titles on pages that already rank is the fastest available gain. The headline guide covers how.
9. Chasing queries AI answers completely
Definitional and simple factual content still has search volume and no longer produces clicks, because the answer appears above the results in full. Writing more of it feels productive and moves nothing.
Keep a small number of these to support your cluster structure, and put the writing effort into the questions where a summary is not enough: comparisons, decisions, costs, processes and anything requiring judgement.
10. A slow site nobody has measured
Speed is rarely the main cause of a traffic problem and it is often a contributing one, particularly on phones. The usual culprits are the same three every time.
Measure it before you optimise it. Half the speed work I see was spent on the part that was already fine.
11. No email capture
Not strictly a traffic mistake, and the reason traffic loss becomes existential. A blog with a list survives an algorithm change; a blog without one starts again from zero.
One clear signup, on every post, saying what you send and how often. That is enough to begin.
12. Quitting at the wrong moment
The last mistake and the one that ends the most blogs. Months three to six look like failure on every metric, and they look identical whether you are on the right track or the wrong one.
Set a review date at thirty published posts, judge it then on impressions and query relevance rather than sessions, and hold the schedule until you get there.
Almost every blog that worked looked like a failure at week twelve. The difference was who was still publishing at week forty.
How to work through this list
In that order, because the first four operate on pages search engines already trust, and the fifth starts from nothing.
Frequently Asked Questions
Check whether it coincided with an algorithm update, whether it is site-wide or page-specific, and whether impressions or only clicks fell. Clicks falling with impressions steady usually means AI answers, not a penalty.
Not for being AI-written. It performs badly when it adds nothing beyond what is already published, which is what most unedited output does.
Consolidate rather than delete where the content has any value, redirecting to the stronger page. Delete only genuinely empty pages, and expect a few weeks of movement afterwards.
Search Console's query view showing two of your URLs for the same query, or a query where your ranking position swings widely week to week.
Title and content changes on pages that already rank often move within two to six weeks. Structural work and consolidation takes closer to three months.
Rarely on a small blog. It is worth ruling out with a crawl, and then the answer is almost always content, structure or headlines.
Before You Go
None of these twelve require a rebuild. Nine of them are edits to pages you have already published, and those are the ones to do first because search engines already know those pages exist.
Start with the titles on your ranking pages, then the cannibalisation, then the updates. If you want the positive version of this list, the writing process and the consistency system between them cover most of it.
Fix the pages you have before writing new ones.
I'm a marketing consultant, entrepreneur and content creator. I help businesses grow through practical marketing, websites, SEO, content and AI.
More About Tariq →