A guest post by Jordan Olivas, founder of Harvv. Harvv and Nina are business partners. Nina isn’t paid for this post.
The short answer
Yes. On a small site, a steady blog can bring a large share of search visits. On sites in our sample with a steady blog (10 or more posts that people keep finding), 40% to 88% of all search visits started on a blog post. On sites with smaller blogs, it was 17% or less.
Two things to know before you judge your own blog. Half of readers spend under a minute on a post, about 200 words. And few buy in the same visit, so measure a post by the search visits it brings and the clicks it sends to the rest of your site.
See the numbers
| Group | Lowest | Highest |
|---|---|---|
| Steady blogs | 40% | 88% |
| Small blogs | under 1% | 17% |
Data: Harvv
Why do some blogs bring most of the search traffic?
The sites where the blog mattered most each had at least 10 posts that people kept finding. That sounds obvious, and it partly is: a post that gets found is, by definition, bringing visits. The useful part is the size of the gap. On those sites the blog was not a side project. It brought 40% to 88% of their search traffic.
The low group includes big brand stores. We didn’t measure why their share was so low. One likely reason is that people search for them by name and land on the home page or a product, so even a busy blog is a small slice. Their blogs still brought real search visits.
Check your own split in Google Search Console: open Pages and add up the clicks on your blog posts against the rest.
How much of a blog post do people actually read?
Less than most writers hope. Across the blog posts we measured, half of readers spent less than a minute on the page (51 seconds). At normal reading speed that’s about 200 words. Only 11% reached the end.
See the numbers
| How far down the post | Readers |
|---|---|
| Halfway | about 50% |
| Three-quarters | 22% |
| The end (90% of the page) | 11% |
Data: Harvv
73% of these readers were on a phone. On the posts where we could compare the two, phone readers dropped off sooner: 18% got three-quarters of the way, against 27% on a computer.
That doesn’t mean you should write short posts. Google says plainly that it doesn’t prefer a word count (Google’s guide to helpful content), and long posts are fine for questions that need them (see Nina’s guide to ideal blog length). It means the answer has to come first, because that’s the part people read.
Do blog readers buy?
Rarely in the same visit. We compared search visitors on the same site, on the sites with enough of both kinds. On one, 0.3% of visitors who landed on a blog post placed an order or sent a form, against 3% of those who landed on any other page. On another, 0.2% against 10%. The blog figures rest on only a handful of orders, so read them as “well under 1%”, not as exact rates.
That’s normal. Someone reading “how to choose” is earlier in the decision than someone on a product page. The useful question is whether the post moves them on:
See the numbers
| Next step | Readers |
|---|---|
| Opened another page | 17% |
| Opened a product page (stores) | 11% |
| Reached the cart (stores) | about 1% |
Data: Harvv
On a store, 11% of readers opening a product page is real work done by the post. But it only happens when the post links to the product, in the text, where the reader is. Nina’s guide to internal linking for blog SEO covers how.
How to write a post that gets found and read
- Answer the question in the first few lines. Half of readers are gone within about 200 words. Google lists no special trick for its AI Overviews (Google: AI features and your website); a clear answer up top serves readers either way. Nina covers how to structure a post for Google AI Overviews.
- Write headings that say what’s in the section. “How much does it cost?” helps a skimmer. “The surprise” doesn’t. In our sample, posts with more headings per 1,000 words saw more readers go on to other pages.
- Add a real picture or chart near the top. We suggest about one per section after that. Posts with at least one image were read further, though they were also shorter, so we can’t say which of the two did it.
- Link each post to the matching product or service page and to two related posts, in the text, with words that say what the page is. Google wants every page you care about linked from another page on your site (Google: link best practices).
- Show who wrote it, and keep dates honest. Google’s own checklist asks whether you change a page’s date to make it look fresh when the content hasn’t really changed (Google’s guide to helpful content).
- Fix the posts Google already shows before writing new ones. A post sitting on page two for a real question often needs a better opening answer and title, not a new sibling.
How we measured this
- A sample of websites that use Harvv: online stores and service businesses of different sizes. They aren’t named, and they’re shown only in two groups.
- Search visits: visits from Google, Bing and other search engines between October 6 and November 2, 2026, with ad clicks and bots removed. A blog post is a page under the site’s blog, news or guides section.
- “Posts people kept finding”: posts with at least 15 visits in those four weeks, from any source. Reading: time and scroll were measured on each page view of those posts. The full reading numbers are in Harvv’s study of average scroll depth.
- “Placed an order or sent a form” counts any order, enquiry or sign-up form. Product page and cart steps apply to the stores only.
- These are patterns across a small sample, not a test. Your site may differ, which is why it’s worth measuring your own.
From Nina
Want a steady blog without writing it yourself?
Nina plans, writes and publishes SEO and AI visibility focused blog posts for your site every month. Strategists pick topics from real search data, Nina’s writing system drafts each post, and U.S.-based editors polish it before it goes straight to your CMS.
