The short answer
About halfway. On the blog posts we measured, the typical reader scrolled 48% of the way down the page. 22% got three-quarters of the way, and 11% reached the end.
Scrolling isn't reading, though. Half of readers spent under a minute on the post (51 seconds), which is about 200 words.
See the numbers
| How far down the post | Readers |
|---|---|
| Halfway | about 50% |
| Three-quarters | 22% |
| The end (90% of the page) | 11% |
Data: Harvv
Data: HarvvWhat is a good scroll depth for a blog post?
If half your readers get halfway down, your post is typical for this sample. That's all a benchmark can tell you. Posts differ in length, layout and what the reader came for, so the useful comparison is one of your posts against your others.
Look for the post where readers stop much earlier than they do elsewhere on your blog. Something near that point is turning them away: a slow image, a wall of text, or an answer they already got.
Does scroll depth show how much people read?
No. It shows how far the page moved. The typical post here ran about 1,150 words, so a reader who got halfway scrolled past roughly 600 of them. But half of readers were on the page for under a minute, time to read about 200 words at a normal pace.
So people skim. They move down the page faster than they read it, looking for the part they need. Put the answer in the first few lines, and write headings that say what each section holds, so a skimmer can stop in the right place.
Do phone readers scroll less?
Yes, in this sample. 73% of blog readers were on a phone. On the posts where we could compare the two, 18% of phone readers got three-quarters of the way down, against 27% on a computer.
See the numbers
| Device | Got three-quarters of the way |
|---|---|
| Phone | 18% |
| Computer | 27% |
Data: Harvv
The same post is a much longer scroll on a phone, so this is partly the screen and not the reader. Either way, most of your readers are on a phone, and what they see is the top of the post.
What do readers do after a blog post?
Most leave. 17% opened another page on the site. On online stores, 11% opened a product page and about 1% reached the cart.
See the numbers
| Next step | Readers |
|---|---|
| Opened another page | 17% |
| Opened a product page (stores) | 11% |
| Reached the cart (stores) | about 1% |
Data: Harvv
Two weak patterns showed up across posts. Posts with more headings for their length saw more readers go on to another page. Posts with at least one image were scrolled further, but those posts were also shorter, so we can't say which of the two did it.
How do you measure scroll depth on your own site?
- Know what Google Analytics counts. Its built-in scroll event fires once, when a visitor reaches 90% of the page (Google: enhanced measurement events). That's the "reached the end" number, not an average depth.
- Record the deepest point of each visit. To see halfway and three-quarters you need extra scroll events, or a tool that records the deepest point reached on every page view.
- Read it next to time on page. Deep scroll with a few seconds on the page is a skim, or a bot.
- Split phone from computer. One blended number hides that most readers see far less of the page.
How we measured this
- 3,891 visits to 75 blog posts on a sample of websites that use Harvv: online stores and service businesses. The sites aren't named.
- The 28 days to November 2, 2026. A post is included when it had at least 15 visits in that time. Bots and the sites' own staff were removed.
- Scroll depth is the deepest point of the page a reader reached on a page view. "The end" means 90% of the page, because the last part is usually the footer.
- Reading time is time with the page open and in use. Words read assumes 240 words a minute.
- The phone and computer comparison uses the 49 posts with enough readers on both, so its two numbers don't average to the 22% for all posts.
- Product page and cart steps are for the online stores only. These are patterns in a sample, not a test.