Straight answer: a good blog post answers the question in the first 100 words, contains at least one real number and one named example, is built to be skimmed in subheads and short paragraphs, and takes a position strong enough that someone could disagree with it. Miss any one of those and the post gets published, sits there, and does nothing.
Most blog posts fail before anyone reads a word of the writing
I’ve read thousands of blog posts as a consultant, and I’ve written a fair few duds myself in twenty-plus years of doing this. The pattern in the bad ones is almost always the same. They open with three sentences of throat-clearing (“In today’s fast-paced digital world…”), they never name a real number, they never name a real thing, and they end with a paragraph that says “in conclusion” and repeats the title back at you.
None of that is a writing problem. It’s a decision problem. Somebody decided it was fine to publish a post that says nothing specific, and hit go anyway.
Here’s the bit most guides on this topic skip, because it’s uncomfortable: a lot of “good blog post” advice is really advice for producing safe, forgettable content. Hit the word count, add three subheads, sprinkle the keyword, done. That checklist will get you a post that Google can crawl. It won’t get you a post anyone remembers, links to, or buys from. The posts that work usually break at least one item on the polite checklist, because they say something specific enough to be wrong about.
A real story: the post that did nothing and the one that brought in clients
A few years ago, when I was rebuilding my business in public after a rough patch, I wrote two posts in the same month on very similar topics: one on why small businesses should be using AI, and one on exactly which three AI tools I’d tested that week and what each one cost me. The first was 1,800 words of the kind of generic “embrace the future of AI” content you can find on a thousand sites. It got 40-odd views in three months and zero enquiries.
The second was 2,300 words, named the tools, gave the actual monthly prices, and included one paragraph where I said plainly that one of the three wasn’t worth the money for a business under 10 people. That post ranked in the top five for its main phrase within nine weeks, and I can trace three consulting enquiries directly back to it because people quoted the specific line about pricing when they emailed me. Same topic, same author, same week almost. The difference was specifics and a stated opinion, not better sentences.
That’s the whole argument of this post in one example. Specificity and a point of view do the work that clever writing gets credited for.
The seven things a good blog post always has
Whatever the topic, length, or industry, I check every post I publish against this list before it goes live. If it fails two or more of these, it goes back for a rewrite, not a polish.
- An answer in the first 100 words. Not a lead-up to an answer. The actual answer, plainly stated, that someone skimming on their phone in a queue could screenshot and use.
- At least one real number. A price, a percentage, a week count, a client figure. “Increased traffic significantly” tells a reader nothing. “Went from 400 to 2,100 monthly visitors in five months” tells them something they can check their own results against.
- One named, specific example. Not “a client of mine” with every identifying detail sanded off. Name the industry, the size of the business, the actual number, even if you change the company name for privacy. Vague anonymity reads as invented, whether it is or not.
- Structure built for skimming. A subhead every 150 to 300 words, paragraphs of two to four sentences, and lists where a list helps rather than where it pads. Most people reading a blog post are scanning, not settling in for a read.
- A stated opinion, not a survey of opinions. “Some experts say X, others say Y” is not a position, it’s a summary. Pick a side and say why, even knowing some readers will push back.
- Something the top-ranking pages on the same topic don’t say. Before I publish, I read the current top five results for the phrase I’m targeting. If my draft says nothing they haven’t already said, it’s not ready.
- A clear next step. Not a hard sell, just one obvious thing the reader can do next, whether that’s a related read, a checklist, or booking a call.
Why length is the wrong question to start with
People ask me constantly what the “right” word count is for a blog post. There isn’t one, and chasing a number backwards is how you end up with padding. I’ve had 700-word posts that answer one narrow question completely and rank fine, and I’ve had 4,000-word guides that needed every one of those words because the topic had that much ground to cover. What matters is whether every paragraph is earning its place. A useful test: if you deleted any single paragraph, would the post get worse? If the answer is no, that paragraph shouldn’t be there.
This connects to something a lot of business owners get wrong about why they’re blogging in the first place. If you’re publishing because a marketing agency told you “you need to blog weekly,” without understanding the real purpose of blogging for a business, you’ll optimise for the wrong thing, usually frequency over substance. I’d rather a client publish one specific post a month than four thin ones a week.
The opening line is doing more work than you think
Readers, and Google, decide within seconds whether a post is worth their time. I test this constantly on my own posts by reading only the first two sentences out loud. If those two sentences could open literally any other post on the same topic, I rewrite them. “In this article, we’ll explore…” could open a million posts. “I lost £3,000 on a Facebook ad campaign in a weekend because I skipped this one step” can only open one.
The same discipline applies to your subheads. If someone read only your subheads, top to bottom, they should get a rough sketch of your whole argument. That’s a fast way to check structure before you’ve even finished writing: list your subheads on their own, read them in order, and see if they tell a coherent story.
Where writers cut corners, and where it shows
The three places I see corners cut most often, in client drafts and in my own early drafts before I fix them:
- The example gets vaguer with every edit. A first draft says “a client in Manchester spent £600 on Google Ads and got 4 leads.” By the third edit, someone’s nervous about specifics and it becomes “a client saw good results with a modest ad spend.” That edit makes the post safer and worse in the same move.
- The middle sags. The opening is sharp, the conclusion is tidy, and the middle third is where the actual information should be and instead there’s filler. That’s usually because the writer knew the hook and the ending before they knew what they had to say.
- Nobody decides what the post is against. Good posts usually push back on something: a common assumption, a piece of bad advice going round, an industry habit that wastes people’s money. Posts with nothing to push against tend to just describe, and description alone rarely gets shared.
If you’re outsourcing this, know what you’re paying for
If you’re not writing the post yourself, the price you pay should reflect most of what’s above, not just a word count. I’ve seen businesses pay £40 for a 1,000-word post that hits none of these marks, and I’ve seen businesses pay £400 for a post that includes a real interview, real numbers, and a genuine stance. If you want a fuller sense of what fair pricing looks like across different levels of quality, I wrote a detailed breakdown on how much it really costs to have blog posts written, and the short version is that cheap word counts and good blog posts rarely live in the same invoice.
It’s also worth deciding early whether the format even fits. Some topics land better as video than text, and forcing a demo or a walk-through into 2,000 words of prose because “we blog” is a habit rather than a decision often produces the flattest posts of all. I’ve broken down when to pick one over the other in the difference between blogs and vlogs and which one pays you back, and it’s worth reading before you commission the next twelve posts on autopilot.
A short checklist before you hit publish
This is the actual pass I run over a draft in the last ten minutes before it goes live, in order:
- Read the first two sentences alone. Do they say anything specific?
- Find the number. If there isn’t one, add one or cut a claim that needed one.
- Find the named example. If it’s vague, either name it or delete it, half-anonymous examples help nobody.
- Read the subheads on their own. Do they tell the story of the post?
- Find the sentence someone could disagree with. If there isn’t one, the post is too safe.
- Check the last paragraph gives one clear next step, not a summary of what they just read.
If you’re just starting out and this feels like a lot to hold in your head at once, start smaller. Even people building a side hustle from a spare room with no marketing background can hit five of the seven marks above on a first attempt, because none of it requires experience, it requires being willing to be specific and take a side.
The uncomfortable part nobody likes admitting
Here’s the truth that sits under all of this and that most advice on blogging quietly avoids: a good blog post will occasionally cost you a reader. Someone will disagree with your number, dislike your opinion, or feel called out by your example. That’s not a sign you did it wrong. Bland, hedge-everything posts that offend nobody also convince nobody, and they get forgotten by lunchtime. I’d rather publish something that a handful of people push back on and a much larger number of people bookmark, than something that a thousand people skim and none of them remember by the following week. If your last five posts haven’t produced a single disagreeing comment, an email, or a “well, ” reply, that’s worth sitting with, because it usually means you haven’t said anything yet.
Frequently asked questions
How long should a good blog post be?
Long enough to fully answer the question and not a word longer. Some useful posts run 700 words because the topic is narrow, others need 3,500 because the topic is broad. Chase completeness, not a target number, and cut any paragraph that could be deleted without the post getting worse.
Does every blog post need a personal story or example?
It needs at least one specific, named example, whether that’s a personal story, a client case, or a documented data point. Generic statements without a specific instance behind them read as unproven, and readers scroll past them without slowing down.
What’s the single biggest mistake people make writing blog posts?
Burying the answer. Most drafts spend the first two or three paragraphs warming up before they say anything useful. Readers and search engines both reward posts that state the point plainly in the first 100 words, then spend the rest of the post proving it.
How do I know if a blog post is good before I publish it?
Run it through a short test: find the real number, find the named example, read the subheads alone to check they tell a coherent story, and find one sentence a reader could reasonably disagree with. If any of those are missing, the post isn’t ready yet, regardless of the word count.
