How to Apply an On Page SEO Checklist to a Blog Post
Six questions to ask about a post that is already written. Not a guide to writing one, since that is a separate subject. The review step almost every business blog skips and the one that catches the faults nobody notices while writing.
This Is A Review, Not A Writing Guide
Writing and structuring a post is covered in our blogging material. This page covers what to check once it exists, before and after publishing, which is a different and considerably shorter exercise.
Why the two are separate. Different questions.
Writing asks what to say and in what order. Reviewing asks whether what was said does the job it was written for. Somebody good at the first frequently never does the second.
Why the review step gets skipped. Publishing feels like finishing.
A post is written, read once for typos and published. That reads as completion. The checks below all happen after the point where everybody stopped.
What a review catches. Faults invisible while writing.
Whether anything links to it, whether it duplicates another post, whether it leads anywhere. None of those are visible from inside the document you are writing.
How long it takes. Minutes.
Six questions, most answerable immediately. The disproportion between the time this takes and what it catches is why the step is worth formalising.
When to run it. Twice.
Once before publishing, which catches the first four questions. Once some months later, which is the only way the last two get answered. Most businesses do neither, though doing the first alone is already most of the benefit.
Why this is not a printable list. Each check needs reasoning.
The questions below carry the reason they matter, then point at the section covering each subject properly. A list of ticks without that produces items nobody can action, which is the failure the audit page describes.
Does It Match One Search
Name the search this post should answer. If you cannot, else if you name two, that is the finding. A post trying to serve several searches serves none of them well.
How to run the check. Say it aloud.
State in one sentence what somebody would type to find this post. Hesitation is the signal, because a post with a clear target produces the sentence immediately.
What two answers mean. Two posts.
A post covering two questions is competing with itself for attention and probably with another post on your site. Splitting it is usually easier than aiming it.
The second half of the check. Does anything else target it.
Whether your site already has a post on this search. That is the commonest thing this review catches on a blog running for years. Cannibalisation covers resolving it.
What a vague answer means. A topic rather than a target.
Naming a subject area rather than a search is the usual response. It indicates the post was written about something rather than for somebody. That is the finding, worth more than any of the checks below.
Why this comes first. Nothing else matters if it fails.
Every other check assumes the post has a target to be measured against. Without one there is nothing to review, only something to read.
Does The Title Describe It Accurately
Read the title without the post. Would somebody know what they were about to get. A title describing something adjacent is the commonest reason a search engine substitutes its own.
What accurate means here. No gap with the content.
The title promises what the post delivers, in the words a reader would use. Not a summary of the subject area. A statement of what this particular post covers.
The common failure. A title written for effect.
Posts frequently get titles chosen to sound appealing rather than to describe. That produces a mismatch, which affects both whether the title is used and whether the visit survives.
What not to check. Its length.
There is no character rule, so measuring it achieves nothing. The meta tags pillar covers why, including what Google has published.
The check behind the check. Whether it was ever rewritten.
Many posts carry the working title they were drafted under, because nobody revisited it once the post was finished. Reading the title against the finished post frequently reveals that it describes an earlier version of the plan.
The related check. The description.
Whether the summary offered for the results snippet exists and matches. Worth a moment on posts people actually reach and not worth writing for posts nobody finds.
Does The Structure Hold Up
Read only the headings, in order, ignoring everything else. If the argument of the post follows from them, the structure holds. If they read as disconnected labels, it does not.
Why the test works. It simulates scanning.
Most readers arrive with one question and scan for it. Reading only the headings is close to what they actually do, which makes it a better test than reading the post properly.
What failure indicates. Usually not the headings.
Vague headings normally mean the post has no clear argument to describe. Rewriting them without fixing the order produces better labels on the same confusion.
What to check about levels. Whether any were chosen for size.
A heading level picked because the text looked better at that size breaks the outline. The headings pillar covers why that matters, including for anybody using a screen reader.
What not to check. How many there are.
No count is correct, so counting them is a check with no possible outcome. Whether they describe the post is the whole question.
Is It Linked To Anything
A post nothing links to is orphaned. It exists, it can be reached by anybody with the address, yet nothing on your site tells anybody it is there. This is the commonest fault on business blogs.
What orphaned means practically. Reachable only from a list.
The post appears on your blog index and nowhere else. Once newer posts push it off the first page, nothing on the site points at it at all.
Why it matters more than it sounds. Links carry weight.
Internal links from related pages are one of the genuinely effective things entirely within your control. A post with none is being supported by nothing.
Why it happens. Linking is a separate action.
Publishing a post does not link to it from anywhere. Somebody has to go back into older pages and add links, which is a second task nobody assigned.
How to run the check. Name two pages.
Which two existing pages should link to this post. If you can name them, the check is whether you added the links. If you cannot, the post may not belong on the site.
What to do about the backlog. Work backwards.
On a blog running for years, most posts are orphaned. Fixing that on the posts that matter is a better use of an afternoon than writing another one.
Does It Point Anywhere
A post that leads nowhere has done its job for the reader and nothing for the business. Somebody finished it, was helped, then left with no idea what you sell or what to do next.
What pointing somewhere means. A relevant next step.
A link to the service the post relates to, else to the next thing somebody would want. Specific to the subject rather than a general invitation to get in touch.
Why this is not conversion advice. It is a link check.
Nothing here is about persuasion or wording. The question is whether a route exists from the post to the part of your site that earns money, which is structural.
What over-doing it looks like. An advertisement.
A post interrupted by requests to enquire stops being useful and stops being read. The reader came for an answer. The next step belongs after they have it.
Why informational posts still need it. The reader is not ready.
Somebody learning about a problem is not going to enquire today. A relevant onward link is what lets them come back to the right place when they are.
Has It Been Checked Since Publishing
Everything above applies at publication. This one applies afterwards, because a post that was correct when written can become wrong without anybody touching it.
What goes out of date. Specifics rather than arguments.
Rules that changed, prices quoted, products discontinued, links to pages that have moved. The reasoning usually survives while the details do not.
What to check. Whether anything is now untrue.
Read it against current reality rather than for quality. A post can be well written and factually stale, which is worse than being badly written and correct.
The link check specifically. Outbound and internal.
Links break silently over years, particularly to other people's sites. A post pointing at pages that no longer exist reads as abandoned.
What not to do. Change the date alone.
Updating a displayed date without revising anything makes the post misrepresent itself. Our blogging material covers doing maintenance properly as a routine.
What Not To Check
Four things routinely checked that are not worth checking. Each appears on published checklists, each is measurable, yet none of them measures anything a search engine uses.
The proportion of the post a phrase occupies. No target exists.
Nothing has ever published one. Every figure in circulation traces back to other guidance. The keywords pillar covers why in full.
How long the post is. Length is not a factor.
Whether the post answers its question completely is the useful question. It is unrelated to how many words getting there took.
How many headings it has. No count is correct.
Whether the headings describe the argument matters. How many exist is a number with no meaning attached to it.
Where the phrase appears beyond the obvious. No map exists.
The title, the main heading and early in the text are sensible because they describe the post. Beyond those, there is no placement rule to check against.
Why these persist. They are easy to measure.
Each produces a number that can be reported and shown to have improved. That makes them appealing deliverables in a way that judgement about whether a post is any good never will be.
Name two pages
that should link
to it.
If you can name them, the check is whether you added the links. If you cannot, the post may not belong on the site. On a blog running for years, most posts are orphaned, so fixing that beats writing another one.
The six checks:
Density, length and heading counts are not on the list, deliberately.
Every guide.
One practice.
What a page says about itself, headings, keywords, search intent, images, page types, user behaviour, credibility and how to review an existing site.
