- Publishing is not distribution. A post with no promotion plan is unfinished, however good the writing is.
- Some post types can never get search traffic. Knowing which ones saves you months.
- Spend as much time promoting one post as you spent writing it. Almost nobody does, and that is the opening.
- Internal links are the cheapest lever on a blog, and the most consistently left out.
- Measure per post, never per blog. A blog average hides the two articles doing all the work.
- A refresh beats a new article most of the time, and lands in weeks rather than quarters.
Why most blog posts get almost no traffic
Open the traffic report on any blog that has been running for two years and the shape is always the same. Two or three articles carry most of the readers, a handful get some traffic, and the long tail — often eighty per cent of everything published out there — gets close to nothing. Teams see that curve and conclude their writing was not good enough. Usually the writing was fine.
The failure is simpler and more fixable. An article was published, shared once, and then left to find readers on its own. Search takes six to twelve months to work out whether a new piece deserves ranking, and it makes that call partly on signals an article only gets if somebody sent it readers first. No promotion means no early engagement, which means no ranking, which means no traffic — a loop that closes quietly and never reopens.
So the honest answer to how to get traffic to your blog starts before the search question. Every post needs a route to its first hundred readers that does not depend on Google, and that route has to be planned at the same time as the outline. Teams that treat promotion as an afterthought spend a year producing work nobody reads, and no volume of writing fixes it.
The rule worth adopting
If a post has no named distribution route before it is written, do not write it yet. That single rule does more for a blog than any amount of keyword research, and it costs nothing to apply.
Not every post can get traffic
Some articles are structurally incapable of earning search readers, and no amount of promotion changes that. Sorting your ideas by type before you commit the time sits among the highest-return habits available, so here is the split our analysts use when they map out a content calendar.
| Post type | Where readers come from | Realistic traffic |
|---|---|---|
| How-to and problem-solving | Search, over time | High, and it compounds |
| Comparison and "X vs Y" | Search, buyers late in the cycle | High, and it converts |
| Original data or benchmarks | Links, then search | High, if the data is real |
| Opinion and commentary | Social, your list | A burst, then nothing |
| Company news and releases | Your list, existing customers | Near zero from search |
| Thin listicles | Nowhere, increasingly | Close to zero |
Read the bottom two rows carefully, because that is where some corporate blogs still spend most of their budget. Company news is worth putting out for customers who already subscribe, but expecting a product release to get search traffic is a category error. It answers a question nobody out in the market is asking.
The top three rows are where a blog earns its keep. They share this: somebody, somewhere, is actively looking for the answer — which means search has a reason to rank the piece and a reader has a reason to click. Everything else is either a favour to your existing audience or a way to keep a publishing schedule looking busy.
What a post that ranks actually contains
Assume you picked a type that can work. The next question is which parts have to be inside the piece itself, and the bar has risen considerably over the past few years. Thin pieces never rank now, so the useful comparison is not against your last article but against the three results currently at the top.
Open those three results and list every sub-question they answer; some of them will surprise you. Your draft needs all of them plus something none of them has — a number, a screenshot, a worked example, an opinion somebody could disagree with. Matching the field gets you considered; adding something nobody else has is what gets you chosen.
A checklist for the piece itself:
- A title that names the outcome rather than the topic. "How to fix X" beats "Thoughts on X" every single time.
- The answer in the first two paragraphs. Readers who have to scroll to find out whether they are in the right place drop out.
- Sub-headings that match the way people phrase the question, so a skimmer can find their part.
- At least something that cost you effort: original data, a real screenshot, a worked calculation.
- Internal links out to two or three related posts and one commercial page.
- A next step at the end. Never "thanks for reading" — something a reader can use.
Notice how little of that is about word count. Long articles rank because they answer more sub-questions, rather than because length is a signal on its own. A tight two-thousand-word piece that answers everything will out-rank a padded four-thousand-word one, and it takes less of your time to produce.
No writing capacity in-house? Our blog content generation service briefs from your own buying questions and ships posts built to this standard.
See how it works →How to promote one post properly
Here is the part almost everybody skips. Use the same amount of time promoting a post as you spent writing it. If the draft took six hours, book six hours for distribution — and book them before publication, because half of what follows has to happen in the first week.
- Email your list. The single highest-response route you own. A short email, a single link, a reason to click. Send it out the day the post goes live.
- Add internal links from older articles. Find three existing pieces on related subjects and link out to the new one from inside the body text. Ten minutes, and it works every time.
- Tell the people quoted in it. If you cited somebody, mention a tool, or used a customer example, let them know. Some of them will share it, and some will link.
- Post where your buyers actually are. Never every platform — just the one or two where your last ten deals started. An article shared into the wrong community gets nothing out of it.
- Answer the question live somewhere. Find the same question being asked in a community or comment thread and answer it properly, using the post as supporting detail rather than the whole reply.
- Send it to five people directly. One-to-one messages to people who will genuinely find it useful out-perform any broadcast, and they occasionally turn into links.
- Add it to a relevant existing page. A resources page, a help centre article, an onboarding email. Somewhere it keeps being seen after the launch week runs out.
- Book a diary reminder for three months out. Re-share it, find out what it ranks for, and decide whether it needs a refresh. Most articles never get this step and quietly decay.
None of those eight steps needs a budget, and together they take less time than writing a second piece. That trade is the whole argument: one post promoted properly will out-perform three published and abandoned, every time, on any blog we have audited.
Which routes work out best
Some routes work out better than others, and the ranking shifts by audience. Use your own data rather than a generic list: pull sessions by referrer for the last three articles you promoted and find out which route actually sent readers. For some teams that is email by a wide margin. For some it is a single community nobody outside the industry has heard of. Very few teams get much out of broad social sharing, and almost none get anything out of paid promotion aimed at a cold audience.
Once you know your top two routes, use them every time and stop pretending the other six matter equally. That is how a small team gets a repeatable launch instead of a scramble, and it tends to increase the readers each piece reaches without adding an hour of work to the blog calendar.
Link your posts to each other
Internal links are the cheapest lever on a blog and the one most consistently left out. They pass some authority between your own pages, they help search work out which post covers what, and they keep readers moving instead of dropping out after a single page. All of that is free and takes minutes.
Use a habit rather than a project here. Every time you publish, add links from two or three older articles pointing at the new one, and add links out from the new one to the pieces it naturally references. Over a year that produces a genuinely connected blog with traffic moving between pages rather than a pile of orphans, and the difference in the way search treats the two is not subtle.
A warning. Links from your navigation or a sidebar widget count for far less than links inside body text, because search can tell the difference between a template and an editorial decision. If a link is worth having, it belongs in a sentence.
Refresh old posts before writing new ones
Any blog older than eighteen months has some articles that used to get readers and quietly stopped. Refreshing those is the fastest route to more traffic, because the article already has links pointing at it and a history search recognises. You are never asking for trust — just for a second look.
Use four moves in a refresh: update the figures, add the sub-questions competitors now cover, rewrite the title, and add internal links from your stronger pages. Keep the URL. Changing it throws out the links and the history, which are the only reasons the refresh works at all — and that mistake turns a cheap win into a loss.
Which articles to refresh first:
- Articles ranking between position six and fifteen. They have some trust already and need more depth.
- Articles whose traffic has fallen more than a third year on year.
- Articles with earned links that send almost nobody onward to a commercial page.
- Anything where the year in the title has aged badly.
Expect roughly three refreshes for the time a single new piece costs, and expect them to land sooner. That ratio is why the answer is usually "update what you have" rather than "publish more", and why it is easier to defend when a budget comes under review.
Measure per post, not per blog
A blog-level traffic number hides everything useful. If two articles out of forty produce most of your readers, the average tells you nothing about which move comes next — and it certainly cannot tell you which post to write more of. So the reporting has to go one level deeper.
| Metric to check per post | Where | What it tells you |
|---|---|---|
| Impressions and average position | Search Console | Whether search has noticed the article yet |
| Click-through rate | Search Console | Whether the title is doing its job |
| Sessions by channel | Analytics | Which promotion route actually worked |
| Onward clicks to a commercial page | Analytics | Whether the post earns its place commercially |
| Links earned since publishing | Ahrefs or Semrush | Whether anybody thought it was worth citing |
Use impressions first, always. They move weeks before sessions do, so an article with rising impressions and no clicks is a title problem you can fix this afternoon — while an article with no impressions at all after three months has a bigger problem and probably needs rewriting rather than promoting.
Some of this you can only get out of the right tool. Use Search Console for anything search-related, because it is the only place that shows the queries your blog appears for. Use your analytics for the channel split. Use a rank tool if you want to find out how your blog compares with a competitor, but treat those absolute numbers as approximate — some tools disagree with each other by a factor of two, and none of them can see your own data.
Do this once a month for the ten articles that matter, never weekly for all forty. Some of them will never justify the attention, and admitting that early frees up the time you use on the good ones.
When your blog is too young to rank
Everything above assumes search will eventually notice you. On a genuinely new blog that assumption does not hold for a while, and pretending otherwise wastes a year. A blog on a domain with no history and no links gets very little patience from search, so early articles have to earn their readers entirely through the routes you use yourself.
That is survivable, and there is a version of the plan that works: fewer articles, more promotion per article, and every early link treated as a priority rather than a bonus. Publishing weekly into silence never works, whatever the algorithm is expected to reward.
Some teams have a harder constraint than patience. If a funding round, a partnership review or a procurement check is booked, the timeline is fixed and no content plan compresses to fit it. Investors open tools like Similarweb during diligence; partners look at whether a site seems alive at all. Those judgements run on numbers a blog cannot move in eight weeks.
There is one honest way to boost the visible numbers while a blog matures, and it belongs in the open rather than left out of a guide like this.
If a date is the constraint rather than the strategy, targeted website traffic is scoped in writing first, with a free audit before anything is agreed.
How delivery works →Two things worth stating plainly about that option. Purchased visits never become customers and never fill in a form, so anybody promising leads from bought sessions is describing something that does not exist. It is also no substitute for the work in this guide — it buys time for your articles to mature, which is a narrow but real use.
For the wider picture, our guides on how to get website traffic and how to increase website traffic cover benchmarks by company stage and the levers that increase traffic outside the blog.
The launch routine, start to finish
Here is the sequence our analysts hand to SMEs, scale-ups and the agencies running a blog for clients. It assumes one owner and a post a month, which is a realistic cadence for a small team.