How to Start a Blog in 2026 (And Whether You Still Should)

Is blogging dead in the age of AI? No. But the game changed, and pretending otherwise will waste a year of your life.

Here is the honest version, before we go any further. The era when you could write three competent "how to" posts a week, sprinkle in some keywords, and harvest a steady stream of Google traffic — that era is mostly over. AI answer boxes now sit at the top of search results and quietly eat the clicks that used to flow down to pages like mine. A lot of thin, search-optimized content has lost the traffic that justified its existence. If your only reason to blog was easy SEO scraps from Google, I won't lie to you: that reason is weaker every month.

And yet I would still start a blog in 2026. So would most serious people I respect. The difference is why — and the why has shifted from "get found on Google" to something more durable: thinking in public, building a reputation, owning a platform nobody can take from you, and producing the one thing AI cannot fake. Let me walk you through it without the hype.

Why still blog, when the robots write essays for free?

Four reasons survive the AI era. None of them depend on Google.

Thinking in public sharpens you. Writing forces you to discover what you actually believe. You think you understand a topic until you try to explain it to a stranger and find the holes. A blog is a gym for your own mind before it is anything for anyone else. I have learned more from the posts I struggled to write than from anything I read that month.

Reputation is leverage. A small, real body of work changes how people treat you. You stop competing on credentials and start getting recruited — for jobs, clients, collaborations, opportunities you can't yet imagine. I wrote more about this in Building a Personal Brand as an African Technologist, and it applies directly here: a blog is the cheapest, highest-leverage reputation engine a normal person can build.

You own the platform. Your blog on your own domain cannot be deprioritized by an algorithm change, demonetized by a policy update, or deleted by a platform that pivoted. That ownership is worth more in 2026 than it was in 2016, precisely because so much of the internet now runs through gatekeepers you don't control.

It generates leads quietly. Not "passive income" in the lazy sense — but a good post that explains something clearly will, for years, introduce you to exactly the kind of people you want to meet. Mine have brought me clients, friends, and one job I never applied for.

Choosing a focus (without trapping yourself)

The advice to "pick a niche" is half right. You do need a center of gravity — a topic readers can associate with your name after reading two or three posts. A blog about everything is a blog about nothing, and a stranger has no reason to come back.

But the rigid, keyword-mapped niche of the SEO era is a trap. It produces blogs that read like spreadsheets. Pick instead a territory — a broad area you genuinely care about — and let your specific obsessions roam inside it. "Software and the craft of building things" is a territory. "Faith and modern life" is a territory. Within those, you can write about anything that grips you, and the through-line is simply you.

Ask yourself three plain questions: What do I know more about than most people I meet? What could I write about for two years without getting bored? What do I wish someone had explained to me? Where those overlap, start there.

Where to publish: your own site versus the platforms

This is the decision people overthink. Here are the real tradeoffs.

Your own website (WordPress, Ghost, Astro, or a static site). Maximum ownership, maximum control, your own domain. The downside is that you are responsible for everything and you start with zero built-in audience. This is what I recommend for anyone serious, and it's cheaper than people assume. If you want the step-by-step on getting online for nothing, I wrote How to Build Your First Website for Free in 2026.

Medium. Easy to start, some built-in readership, decent for essays. But you're renting. Your work lives behind their paywall and their design, and their distribution can turn off overnight.

Substack. Excellent if a newsletter is your real product — it bundles publishing, email, and payments. The catch is that it's optimized for email subscribers, not search discovery, and again, it's their house.

Hashnode or dev.to. Strong if you write for developers. Built-in technical community, and Hashnode lets you map your own domain, which softens the ownership problem.

Write what AI can't

This is the whole game now, so I'll be blunt. A language model can produce a competent, forgettable explainer on any topic in nine seconds. If your blog post is also a competent, forgettable explainer, it has no reason to exist. The bar moved.

What the machine cannot do is be you, on a specific Tuesday, having actually done the thing. It cannot tell the story of the migration that broke production at 2 a.m. and what you learned crawling out of it. It has no scar tissue, no opinion it would defend over dinner, no embarrassing failure it's still slightly bitter about. It has read everything and lived nothing.

So write the things that require having lived: real experience, earned opinion, specific stories, the messy how-it-actually-went. Take a position. Be willing to be wrong in public. The most valuable sentence on the modern internet is "Here is what happened to me, and here is what I now think because of it." No model can write that sentence honestly.

Consistency beats brilliance

Most blogs die not from bad writing but from silence. The author posts furiously for three weeks, burns out, and the site becomes a tombstone with a 2024 date on the last entry.

The fix is humbler than it sounds: pick a cadence you can sustain on your worst week, not your best. For most people starting out, that's one post every week or two — not three a day. A blog with twenty thoughtful posts published over a year, steadily, will outperform one with sixty rushed posts published in a frantic month and then nothing. Compounding only works if you don't stop.

Be realistic about timelines, too. Meaningful readership takes months, sometimes a year or more. Anyone promising otherwise is selling something. The good news is that the work compounds in your own head from week one, so you're never writing for nothing while you wait.

The SEO that still works (and the kind that doesn't)

SEO isn't dead — but the cynical version is. Stuffing keywords into thin content to game the algorithm now competes directly with an AI that does that faster and for free. You lose that race.

What still works in 2026 is the unglamorous foundation. Write genuinely useful, genuinely original content — the kind a person bookmarks and a model can't replicate. Cover topics with real depth so your page is the best answer, not the fortieth adequate one. Use clear titles and structure so both humans and machines understand what you're offering. Earn links the old way, by being worth linking to. And increasingly, write to be cited by AI answers and recommended in conversations, not just ranked in a list of blue links. The center of gravity is shifting from "rank #1" to "be the source the answer trusts."

Own your audience with a newsletter

Here's the lesson that took me too long to learn: traffic is rented, but a subscriber is owned. Search rankings rise and fall. Social platforms decay. But an email list is a direct line to people who chose to hear from you — and it moves with you no matter what any algorithm does next.

So treat your blog and your newsletter as one system. The blog earns attention; the newsletter captures it. Every post should make it easy and appealing to subscribe, because a reader who enjoyed one post and left is a stranger again tomorrow, while a subscriber is a relationship. In an AI-answers era where discovery gets harder, the people who already said "yes, email me" are the most valuable asset you can build.

How-to

How to publish your first blog post this week

A realistic, low-pressure path from zero to your first published post in seven days — without overthinking the tools or chasing perfection.

Estimated time: P7D

1. 01

   Pick one topic you've actually lived

   Choose a single thing you know firsthand — a problem you solved, a lesson you learned, a thing you wish someone had explained to you. Not a keyword. One specific, real topic.

2. 02

   Choose the simplest platform you'll actually use

   Don't agonize. A free site on your own domain is ideal, but a Medium, Substack, or Hashnode account you'll publish to today beats the perfect setup you never finish. You can migrate later.

3. 03

   Write a messy first draft in one sitting

   Open a blank document and tell the story or explain the thing as if to a friend. Do not edit yet. Aim for honesty and specificity over polish. Eight hundred words is plenty.

4. 04

   Cut, sharpen, and add one real opinion

   The next day, reread it. Delete the throat-clearing. Make the title clear and specific. Add at least one sentence stating what you actually think — the part an AI would never risk.

5. 05

   Add a way to subscribe

   Place a simple newsletter signup or follow link near the end. This single step turns a one-time reader into an audience you own. Don't skip it.

6. 06

   Publish before it's perfect, then share it once

   Hit publish. Send it to three people who'd find it useful and post it in one place you already exist online. Then start thinking about next week's post. Momentum matters more than this one post's reach.

A realistic word on effort

I want to leave you with the truth rather than the pitch. Blogging in 2026 is more work for less immediate reward than it was a decade ago. The easy traffic is gone. You will write posts that almost nobody reads. You will wonder if it's worth it somewhere around month three.

But the people who keep going — who write because it sharpens their thinking and builds something that is unmistakably theirs — end up with an asset no platform and no model can take from them. That was always the real prize. The Google traffic was just a bonus we got addicted to. Build for the prize.

Frequently asked questions

Is blogging dead in 2026?

No, but the old reasons to blog are weaker. Easy SEO traffic has declined as AI answer boxes capture clicks. Blogging is very much alive as a tool for thinking in public, building reputation, and owning an audience — just not as a quick traffic-and-ad-revenue scheme.

Should I start on my own site or on a platform like Medium or Substack?

If you're serious, publish the canonical version on your own domain so you truly own it, and optionally syndicate to platforms for reach. If you just want to start writing today with zero friction, a platform is fine — you can always migrate the content later.

How long until a new blog gets meaningful traffic?

Realistically, months — often a year or more for steady readership. Anyone promising fast results is selling something. The upside is that the thinking and reputation benefits start from your very first post, so the wait isn't wasted.

Does AI make my writing pointless if it can write the same article?

Only if you write what AI writes — generic explainers. Lean into what models can't fake: your real experience, your specific stories, and your honest opinions. 'Here is what happened to me and what I now think' is a sentence no model can write truthfully.

How often should I post?

Pick a cadence you can sustain on your worst week, not your best — for most beginners that's once a week or once every two weeks. Consistency over years beats a burst of posts followed by silence.

Further reading on this site

• How to Build Your First Website for Free in 2026 — the practical companion to this piece, for getting your blog online without spending a cent.
• Building a Personal Brand as an African Technologist — why reputation is leverage, and how to build it honestly.
• Browse Creator Economy — more essays on writing, audience, and building things that are yours.

If this was useful, subscribe to the newsletter — it's the best way to own the relationship with readers I keep arguing for, and the easiest way to get the next piece.