- The post that taught me this the hard way
- The skeleton that keeps people reading
- The paragraph rule nobody follows (until they're forced to)
- Where bullet points help and where they quietly hurt
- The uncomfortable bit most formatting advice skips
- A step-by-step for formatting before you hit publish
- Format changes depending on where the post lives
- Images, white space, and the parts people forget
- Blogging itself is still worth the effort
- What good format cannot fix
- Frequently asked questions
The short version: a blog post that holds readers has short paragraphs (two to three sentences), a subhead every 150 to 300 words, one clear idea per section, and white space doing half the work. But the format only matters if you're writing for the reader who finishes the post, not the one who scans for ten seconds and leaves anyway, and most engagement advice ignores that split entirely.
The post that taught me this the hard way
In 2019 I published a piece on personal branding. Four thousand two hundred words, one long block after another, no subheads until the conclusion. I was proud of it. It was thorough, it was well researched, and it sat on the page like a brick wall.
Google Search Console told me the truth I didn't want to hear. Average time on page was 42 seconds. Bounce rate was over 80%. I hadn't written a blog post, I'd written an essay and dumped it online, and nobody was going to wade through it on their phone during a coffee break.
I rewrote it three months later. Same research, same points, same length roughly. But I broke it into 11 sections with a subhead on every one, cut every paragraph down to two or three sentences, added four bullet lists and two short numbered steps. Time on page jumped to 3 minutes 10 seconds. Organic traffic to that URL tripled over the following six months. Nothing about the substance changed. Only the shape did.
The skeleton that keeps people reading
Here's the structure I use on almost every post now, built from what worked on that rewrite and dozens since:
- An opening line that answers the question in the title within the first two sentences, no throat clearing
- A short paragraph (one or two lines) that sets up why this matters right now
- Subheads every 150 to 300 words, each one a mini promise of what the next chunk covers
- At least one bulleted or numbered list by the halfway point
- A concrete example, a number, or a step-by-step in every major section, never just opinion
- A short closing section that tells the reader what to do next
Notice what's not on that list: a "hook" gimmick, an emoji in every heading, or a forced question every third line. Those tricks work for a scroll on social media, not for someone who arrived from Google with an actual problem to solve.
The paragraph rule nobody follows (until they're forced to)
Two to three sentences per paragraph. That's it. Not because short paragraphs are magic, but because on a phone screen a five sentence paragraph looks like a wall even when it's only 80 words. I check this myself by opening the draft on my phone before publishing, not on my laptop, because that's where most of my readers are reading it.
Roughly 60 to 65% of blog traffic on my own site comes through mobile devices, and that number has climbed every year I've tracked it. If your format only looks good on a 27 inch monitor, you're formatting for yourself, not for your reader.
Subheads should do more than break up text
A good subhead is a mini headline. It should make sense on its own if someone is scanning down the page deciding whether to keep reading, which most people are, at least at first. I write my subheads last, after the section is done, because it's much easier to summarise a point that already exists than to write a heading and then hope the content matches it.
Where bullet points help and where they quietly hurt
Bullets are brilliant for lists of options, steps, or comparisons. They're terrible for anything that requires nuance or sequence with cause and effect, because they flatten everything to the same weight. I've seen posts turn a interesting argument into seven bullet points that read like a shopping list, and the argument disappears along with the paragraphs.
My rule: if removing the connecting words changes the meaning, it needs a paragraph, not a bullet. If the connecting words are just "and", it can probably be a list.
The uncomfortable bit most formatting advice skips
Here's what nobody wants to say out loud. A huge share of the "engagement tips" floating around are built for the reader who was never going to convert, buy, subscribe, or remember your name anyway. The ten second scanner who bounces regardless of how you format the page.
Chasing that reader with shorter and shorter paragraphs, more and more subheads, and endless bolding eventually strips the substance out of your writing until there's nothing left worth reading for the 15 to 20% of visitors who do read the whole thing, and those are usually the ones who email you, share the post, or become a client. I've watched clients rewrite useful posts into skimmable mush chasing a lower bounce rate, and lose the depth that made the post worth linking to in the first place.
Format for skimming near the top of the post so people can tell within ten seconds this is relevant to them. Then write for the reader who stays. Both things can be true on the same page.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
A step-by-step for formatting before you hit publish
- Read your opening two sentences alone. Do they answer the title? If not, rewrite them.
- Count your paragraphs. Anything over four sentences, split it or cut it.
- Add a subhead everywhere the topic shifts, roughly every 150 to 300 words.
- Find at least one place for a list, a table, or a numbered set of steps.
- Check on your phone, not your desktop, before publishing.
- Read the last paragraph out loud. It should tell the reader exactly what to do next, not just summarise what they already read.
Format changes depending on where the post lives
A long-form blog post and a LinkedIn post are not the same animal, and I still see people copy and paste one into the other and wonder why it flops. LinkedIn strips out most rich formatting, so markdown formatting on LinkedIn behaves very differently to what you'd do on a WordPress post, and line breaks do most of the heavy lifting there instead of headers. If you're regularly short on what to say between long posts, running out of angles is its own problem, and I've written before about finding content ideas for LinkedIn when the well runs dry.
Before you even open a blank document, it's worth checking whether the topic is worth the structure at all. I've made the mistake of building a beautifully formatted 2,000 word post around a keyword that had almost no search demand, which is why I now check volume more than once, something I go into in how many times you should check keyword search volume before writing. Format is wasted effort on the wrong topic.
Images, white space, and the parts people forget
An image or a visual break roughly every 300 to 400 words gives the eye somewhere to rest, especially on longer posts. It doesn't need to be a stock photo. A screenshot, a simple chart, a pull quote in a different visual weight, all of that works. What matters is that the page doesn't feel like one continuous grey block from top to bottom.
White space is not wasted space. Margins, line spacing, gaps between sections, they all reduce the mental effort of reading, and reduced effort is exactly what keeps someone scrolling instead of bouncing.
Blogging itself is still worth the effort
None of this matters if you've already decided blogging isn't worth doing for your business, so it's worth settling that question first. I laid out the actual case for and against it in what blogging is and its pros and cons for business, and format is what turns a blog from a nice-to-have into something that pulls its weight. If you're weighing up blogging against video entirely, that's a separate decision covered in blogger versus vlogger, but the formatting rules above apply just as much to a video script as they do to a page of text.
What good format cannot fix
Format buys you attention. It doesn't buy you substance. I've seen beautifully structured posts with short paragraphs, perfect subheads, and clean bullet lists that said absolutely nothing new, and readers can feel that even if they can't name it. They leave anyway, just a bit later than they would have with a wall of text. Format is the container, not the content, and no amount of white space rescues an argument that doesn't hold up.
Tools can help you check this faster than you'd think, and it's worth knowing what's changing in that space rather than assuming your current tool stack is fixed forever, which is why I keep an eye on things like the next ChatGPT update and whether it's worth caring about before I recommend new workflows to clients.
I built a free calculator for your engagement rate for exactly this.
Frequently asked questions
How long should paragraphs be for better blog engagement
Two to three sentences on average, occasionally one line for emphasis. Anything longer than four sentences should be split, especially since most readers are on a phone screen where a long paragraph looks even denser than it is.
How often should you use subheads in a blog post
Roughly every 150 to 300 words, or wherever the topic shifts. Each subhead should work as a mini summary on its own, useful to someone scanning down the page before they commit to reading a section fully.
Do bullet points improve engagement
They improve scannability for lists, steps, and comparisons, but they flatten nuance. Use them for options and sequences, and keep paragraphs for anything where the connections between ideas matter as much as the ideas themselves.
What's the biggest formatting mistake bloggers make
Writing the whole post for the reader who skims for ten seconds and leaves regardless. That reader deserves a clear, scannable opening, but the readers who stay and convert need real depth further down, and stripping that out to chase a lower bounce rate usually backfires.