How to Get Traffic to a New Blog
A new blog gets almost no search traffic for its first three to six months, and this is normal rather than a sign that something is broken. Google needs time to crawl, index and decide where a new domain belongs, and no amount of technical work shortcuts that.
This article uses real measurements from this site — 89 articles, 486 indexed pages, and 26 clicks from 90,700 impressions in its first month — to show what the early months actually look like, and which of the standard advice measurably moved the numbers.
Almost every article on this topic is written by someone selling a course, using numbers from a site that succeeded years ago under different conditions. This one uses figures from a site that is currently three months old and still early — including the parts that did not work.
The actual numbers
CalcFern, the site you are reading, at roughly three months old:
| Metric | Value |
|---|---|
| Published articles | 89 |
| Total words written | ~150,000 |
| Working tools built | 327 |
| Pages indexed by Google | 486 |
| Pages not indexed | 332 |
| Impressions, first full month | 90,700 |
| Clicks, first full month | 26 |
| Click-through rate | 0.03% |
| Average position | 52 |
Twenty-six clicks. Not twenty-six thousand. For 150,000 words and 331 tools.
Nothing was broken. Every page returned correctly, the sitemap was processed, 486 pages were indexed, and there was no penalty. The site was simply new, and Google had no reason yet to rank it above established competitors.
If your blog is three months old and getting nothing, this is what that looks like from the inside. It is the normal shape of the curve, not evidence of failure.
Impressions are not traffic
The most misleading number in Search Console. 90,700 impressions sounds like an audience; it produced 26 visits.
An impression means your page appeared somewhere in results — including position 52, which is page five or six. Nobody scrolls there. Of those 90,700 impressions, the overwhelming majority were at positions where being listed is worth nothing.
Watch clicks, not impressions. A rising impression count with flat clicks usually means you are ranking deeper for more queries, not building an audience. It feels like progress and is not.
The finding that changed the whole strategy
This is the measurement worth taking away, and almost nobody publishes it.
Grouping queries by type and comparing click-through at page-one positions produced this:
| Query type | Top-10 impressions | Clicks | CTR |
|---|---|---|---|
| Unit conversions ("4.5m to ft") | 1,771 | 3 | 0.17% |
A normal result at position 7 earns somewhere between 3% and 8%. Getting 0.17% while ranking on page one is not a title problem or a content problem. It means Google answered the question itself, in a calculator widget above every organic result. The user got their answer and never scrolled.
The lesson generalises far beyond unit conversion. Before targeting a keyword, search it yourself and look at what sits above the organic results. If Google shows a widget, a direct answer box, or fills the screen with its own panels, ranking first there may still earn you nothing.
Queries where Google cannot answer for you — troubleshooting, multi-step procedures, comparisons, anything requiring judgement — are where a ranking converts into a visit.
Indexed is not the same as crawled
Search Console reported 486 pages indexed and the sitemap processed successfully, which looks like everything is fine. The server logs told a different story.
Of 434 URLs in the sitemap, only 147 had been crawled by Googlebot in a two-week window. And crawl frequency had fallen from about 153 hits a day to 17.
Splitting those pages by whether anything linked to them:
| Page had… | Crawled in the window |
|---|---|
| At least one internal link pointing to it | 50% |
| Nothing linking to it (orphaned) | 29% |
Pages nothing linked to were crawled at roughly half the rate. Google finds a URL in your sitemap, but a page with no internal links reads as unimportant, so it waits in the queue — which is what "Discovered – currently not indexed" means in the Search Console page indexing report.
Google's own documentation on how Search works is explicit that crawling, indexing and ranking are three separate stages. A new site can pass the first two and still get nothing from the third, which is exactly the state described above.
If you have server access, check your own logs. Grep for Googlebot and count distinct URLs. It is the only way to see what Google is actually fetching rather than what it says it knows about, and the gap between the two is often large.
"Not indexed" is not the same as broken
Search Console's page indexing report on this site reads 486 indexed, 332 not indexed. The second number looks alarming and mostly is not.
That report defaults to All known pages — every URL Google has ever seen on the domain, including ones deliberately removed. This site cleaned up 353 low-quality pages earlier in the year and redirected each one to a better page. Google still remembers all 353, and lists them as not indexed with the reason Page with redirect.
That is the redirect working correctly. It is not an error, and there is nothing to fix.
Check the reasons before worrying about the number. Expand the "Why pages aren't indexed" list and sort by what they actually mean:
| Reason | What it means |
|---|---|
| Page with redirect | Normal. The old URL redirects somewhere better. |
| Alternate page with proper canonical tag | Normal. You told Google which version to keep. |
| Excluded by 'noindex' tag | Normal if deliberate; check if not. |
| Not found (404) | Fine for deleted pages, a problem if something is linking to them. |
| Crawled — currently not indexed | Worth attention. Google looked and chose not to index. Usually a quality or duplication signal. |
| Discovered — currently not indexed | Worth attention. Google knows the URL but has not crawled it yet — often a crawl-budget or internal-linking problem. |
Only the last two rows are worth acting on. The fix for "Discovered" is usually internal links, which is the same lever described above. The fix for "Crawled — currently not indexed" is usually that the page is thin, near-duplicate, or does not answer anything a person searched for.
To see only the pages you actually care about, switch the dropdown from All known pages to All submitted pages. That filters to URLs in your sitemap — the ones you chose to publish — and is a much more honest measure of whether indexing is working.
What measurably helped
Ranked by what actually moved something, not by what is commonly recommended.
1. Internal links to every page
Fixing 40 orphaned articles — giving each an inbound link from a page that was already being crawled — directly addressed the 29%-versus-50% gap above. This is the highest-leverage thing available to a new site, because it costs nothing and you control it entirely.
Every page should be reachable from another page, not only from your sitemap and archive listings.
2. Removing thin pages, not adding more
The site originally had 396 published posts. Most were templated and near-identical, and they were cut to 55, with proper redirects. That felt like destroying the site. Impressions did fall — but average position improved from 52 to 38, and click-through went from 0.03% to over 14%.
Fewer, better pages beat more, thinner ones. Volume of content is not the lever people assume.
3. Writing for queries with no instant answer
Following the CTR finding above: troubleshooting and how-to content, where a searcher must click to get the answer. This is a strategy change rather than an optimisation, and it is the one most likely to matter over the next year.
4. Titles and descriptions that fit
Google truncates titles around 60 characters and descriptions around 155. Twenty-seven descriptions on this site were over the limit — some at 229 characters — and were being cut mid-sentence in results. That directly suppresses clicks on impressions you have already earned, which makes it one of the few fixes with immediate effect.
What did not help
- More technical SEO, past a point. Once the audit was clean — no thin pages, no orphans, correct schema, valid sitemap, mobile-friendly — additional markup changed nothing. Technical SEO removes obstacles; it does not create demand.
- Publishing volume. Twenty new articles in a fortnight produced no measurable traffic change in that period. Content compounds over months, not days.
- Submitting to IndexNow. Useful for Bing, Yandex and Seznam. Google does not use IndexNow at all, so it does nothing for the search engine that matters most to nearly everyone.
- Chasing average position. That number averages across every query you appear for, including hundreds you rank 80th for. It can only be "improved" by appearing for fewer things — which means less traffic, not more.
A realistic timeline
| Period | What to expect | What to do |
|---|---|---|
| Months 1–3 | Near-zero clicks. Impressions may look large and mean nothing. | Publish. Fix technical basics once. Do not panic. |
| Months 4–6 | First consistent visitors on long-tail queries | Notice which pages moved and write more like them |
| Months 6–12 | Traffic becomes measurable; some page-one rankings | Improve existing pages as well as adding new ones |
| Year 2+ | Compounding, if you kept publishing | Earn links; that becomes the ceiling |
The single biggest predictor of whether a blog works is whether it is still being published in month six. Most are abandoned in month three — precisely when the curve is about to start.
The constraint nobody can write around
After all the technical work, the honest bottleneck for a new site is authority: whether other sites reference yours. No amount of on-page optimisation substitutes for it, and it is the one thing content alone cannot manufacture.
Genuinely useful contributions in places where your work answers a real question — a forum reply, a relevant directory, a comment that helps — do more than another twenty articles. That is unsatisfying advice because it is slow and cannot be automated, which is exactly why it still works.
Traffic is the prior problem, but if you are wondering what it is worth once it arrives, how blogs actually make money covers the five models, the traffic each needs, and why ad rates differ so sharply by country.
The check described above — searching a query to see whether Google answers it itself — is the core of choosing topics. How to find blog post ideas covers where the topics come from in the first place, and how to filter them.
That authority problem is the one worth working on deliberately: how to get backlinks covers what actually earns links for a small site, and which tactics put the site at risk.
If you are not sure whether your pages are indexed at all, why a website is not showing up on Google works through the diagnosis and the blockers in order.
Frequently asked questions
How long does it take for a new blog to get traffic?
Three to six months before consistent search visitors, and six to twelve before it is meaningful. This site published 89 articles and 331 tools, and earned 26 clicks from 90,700 impressions in its first month indexed. That is the normal shape for a new domain.
Why does my site have impressions but no clicks?
Usually because those impressions are at positions nobody scrolls to — position 40 or 50 counts as an impression and earns nothing. If you have impressions with no clicks at page-one positions, check whether Google answers the query itself with a widget or answer box above the results.
Is it normal to get zero traffic in the first months?
Yes. Google needs time to crawl, index and evaluate a new domain, and it has no history to judge you on. Provided pages return correctly, the sitemap is processed and there is no manual action, near-zero early traffic is expected rather than diagnostic.
How do I know if my pages are actually being crawled?
Search Console's Crawl Stats shows totals, but server access logs show exactly which URLs Googlebot fetched. On this site only 147 of 434 sitemap URLs had been crawled in a two-week window — a gap invisible from Search Console alone.
Does publishing more articles increase traffic faster?
Not in the short term. Twenty articles published in a fortnight produced no measurable change in that period. Content compounds over months, and fewer strong pages outperform many thin ones — cutting this site from 396 posts to 55 improved average position and click-through.
Should I submit my site to IndexNow?
It is free and helps Bing, Yandex and Seznam, so there is no reason not to. But Google does not use IndexNow, so it will not affect the traffic source that matters most. For Google, request indexing manually in Search Console.
What is the fastest way to get traffic to a new blog?
There is no fast way through search. The quickest legitimate wins are making sure every page has internal links pointing to it, targeting queries Google does not answer itself, and getting a few genuine references from other sites. Everything else is patience.
Conclusion
The first months of a new blog look like failure from the inside, and mostly are not. Impressions without clicks, pages indexed but not crawled, and rankings that earn nothing are all normal early states.
What you control: link internally to everything, keep pages substantial rather than numerous, and target queries where a click is the only way to get the answer. What you cannot control is time — and after three months, that is the honest answer to why the numbers are what they are.
Related: choosing a blogging platform covers where to publish, how long a blog post should be covers the length question, and making a site load faster covers the technical basics worth doing once.
Comments