Bounce rate: how to calculate it and genuinely reduce it

Bounce rate: what it is and how it is calculated

Bounce rate is the share of sessions in which a user landed on a page and ended the visit without any interaction. The formula is simple:

bounce rate = (sessions without interaction / total sessions) × 100%

The key word is "interaction", because the result depends on how it is defined. In Google Analytics 4, a session stops being a bounce when the user scrolls the page at least once, clicks an outbound link, watches a video, downloads a file, or spends a set amount of time on the page. In Universal Analytics, a session without interaction meant arriving and leaving without clicking anything. This difference means that numbers from GA4 and the old UA are not comparable, and migrating reports without adjustment can trigger a false alarm.

The second problem is interpretation. A session without interaction does not always mean failure. Imagine a customer types your company name into Google, lands on the homepage, sees the phone number, and calls. In GA4 that session will be marked as a bounce, even though it ended in a valuable contact. From a sales perspective, that is a valuable contact. That is why bounce rate has to be read together with the page type and traffic source, not as a single verdict on the whole site. If you want to organize measurement across the entire website, it is worth starting with a review of the structure and goals of individual pages, which we cover in the context of a website project for B2B companies.

It is also worth remembering that bounce rate is a backward-looking metric. It tells you what happened, but not why. In analytics there are indicators that look to the future, such as CLV, the value of a customer over the entire relationship. In the book "Analytics" (2.5 Customer Analytics), the authors point out that classic RFM metrics and similar ones "do not take into account future potential of a customer", while only CLV "directly accounts for future value". That is a good analogy for bounce rate: it shows the past, so do not base decisions about your entire strategy on it.

Where a good result comes from and where a bad one does

Five groups of factors affect the result. The first is user intent. Traffic from a navigational query (your company name) will almost always produce a high bounce rate, because the user came for contact details, not for content. Traffic from a problem-based query ("how to calculate bounce rate") will behave the opposite way.

The second group is how well the content matches the promise. If the headline and meta description announce a calculator and the user lands on a general article, they leave within seconds. The third is page speed and stability. Slow loading on mobile is the most common cause of bounces that nobody notices in reports, because they look like "normal" traffic.

The fourth group is page layout. No visible next step, a wall of text without subheadings, a newsletter pop-up in the first second. The fifth is traffic quality: bots, low-intent traffic from an ad network, misconfigured campaigns sending users to the homepage instead of a service page.

A bad result rarely has a single cause. It is usually the sum of small things: the page loads in 4 seconds, the headline does not answer the question from the search query, and the only button leads to a contact form. Each of these elements on its own lowers the result by a few percentage points, and together they can push it above 80%.

Realistic bounce rate ranges for B2B companies

There is no single safe number. Ranges differ between industries, page types, and traffic sources, and trying to reduce them to one threshold leads to wrong decisions. Below are the typical ranges we observe in B2B projects. These are estimates based on agency experience, not hard data from a single industry study, so treat them as indicative only, not as rigid norms.

For comparison: e-commerce sites and media outlets show different results because the intent is different. In B2B the decision cycle is longer, and users often return to the site several times before filling in a form. So a first session with a high bounce rate does not have to mean a lost lead. Instead of comparing yourself to generic benchmarks from the internet, build your own reference base: record results for each page and track changes over time. A rise from 55% to 70% on the same page is a signal you need to act on. The number 70% on its own, without context, tells you nothing.

How to reduce bounce rate: concrete actions

Start with segmentation, not a redesign. Split the report by traffic source and page type. Check mobile and desktop separately. Only then pick two or three pages with the highest rate and the most traffic. Optimizing a page that gets 30 visits a month will change nothing.

Improve loading speed

This action delivers the highest return. Check Core Web Vitals, especially LCP and INP. Image compression, lazy loading, limiting external scripts, and enabling caching are changes you can implement in days, not months. On mobile, every second of delay increases bounces.

Match content to the search query

Take a page and check which queries bring traffic to it. If the user is searching for "how to reduce bounce rate" and the page talks generally about "conversion optimization", the mismatch is obvious. Rewrite the H1 and the first paragraph so they answer the search question within the first two sentences.

Add a clear next step

Every page should have one main call to action and one secondary, lower-commitment option. For B2B a good combination is "talk to us" plus "download the checklist". A user who is not ready to get in touch stays on the page longer and comes back later.

Rework the page layout

Short paragraphs, subheadings every 2-3 paragraphs, a list instead of a wall of text, key sentences highlighted. Readers scan, they do not read. If they do not find an answer in the first 5 seconds, they leave.

Check traffic quality

Filter out bots, review ad campaigns to make sure they point to the right pages, and verify that links in emails and social posts lead to specific content rather than the homepage. Part of a high bounce rate is simply misdirected traffic.

After implementing changes, wait at least two weeks before drawing conclusions. A shorter period gives you noise, not signal. Compare the same traffic type and the same page so you do not mix the impact of seasonality with the effect of the change.

How to implement changes without burning through your budget

Not every company has a team that can handle analytics, content, and technical work at the same time. In practice, what is usually missing is one person who connects data with decisions. In such situations, a sensible solution is an interim marketing manager, someone who comes in for a set period and is accountable for the outcome, not for hours in the calendar. They work on your data, prioritize changes, and pass knowledge on to the team.

If bounce rate is one of several metrics that need to be sorted out, it is worth looking at your marketing efforts as a whole. Bounce rate rarely improves in isolation; it usually goes hand in hand with better content-to-audience fit and a consistent message at every stage of the funnel. We cover this in more detail as part of B2B marketing strategy.

A practical plan for the first month: pick three pages with the highest traffic and the highest bounce rate, check loading speed, content alignment with the query, and the visibility of the call to action. Make the fixes, measure results after two weeks, and repeat the cycle on the next pages. In most cases, this is enough to see movement in the right direction without rebuilding your entire site.