Website traffic statistics: the practical guide
Five statistics are enough to steer a website: number of visitors, traffic sources, page views, session duration and conversion rate. On their own, they mean nothing. Cross-referenced, they answer the only question that matters to an owner: is this traffic bringing in customers? Google Analytics 4 and Search Console cover the essentials for free. The rest, endless spreadsheets and paid tools included, mostly serves to waste time or watch the competition. Here is how to read these numbers to decide,
Key takeaways
- A unique visitor is not the same as a visit: one person can generate several sessions in a month, which explains the gap between the two figures.
- Direct traffic often exceeds 30% of sessions, but a large share actually hides mislabeled sources (dark social, emails, apps).
- Average session duration hovers around 2 to 3 minutes on a brochure site: above or below that, you need to look page by page.
- Bounce rate alone means nothing: always cross-reference it with the landing page and the source to know whether it is good or bad.
- Google Analytics 4 and Search Console cover an SMB's core needs for free; third-party tools only serve to spy on competitors.
Contents
- What traffic statistics really measure
- The 5 statistics to track first (and which to ignore)
- Where your traffic comes from: decoding sources
- Reading duration, page views and bounce rate correctly
- Linking traffic to results: conversions and ROI
- How to find out a website's traffic: the tools
- Building a simple dashboard and reading it every month
- Taking back control of your numbers without burning your evenings
- Frequently asked questions
What traffic statistics really measure
Most business owners confuse three concepts their tools display side by side: visitors, visits and sessions. This confusion skews every reading. A site can show 800 "users" and 1,400 "sessions" in the same month without any error. These are two distinct realities.
The same person coming back three times in a month counts as a single visitor, but three visits. Logical, when you think about it. You wouldn't say a loyal customer who drops by your shop every week is a "new customer" each time. Data works the same way. This gap matters a lot: a high visits-per-visitor ratio signals loyalty, not necessarily more awareness.
Visitors, visits, sessions: what's the difference?
A visitor is a unique person. A visit (or session) is a stint on the site. A session groups all the pages viewed in one go, up to 30 minutes of inactivity under Google Analytics 4's default definition.
Take a tradesperson whose brochure site gets 500 unique visitors in a month. If each one returns 1.3 times on average, that adds up to about 650 sessions. The session count climbs faster than the visitor count on sites where people hesitate before getting in touch, like construction or B2B services. It's a sign of deliberation, not a flaw. Google details these concepts in its official Google Analytics help on metric definitions.
Why your tools never give the same number
Two tools almost always show different numbers for the same site. Three factors explain it: rejected cookies, data sampling and the very definition of a session. Not a bug. Structural.
Since consent rules came into force, every visitor who declines cookies becomes partly invisible to Google Analytics 4. France's data authority, the CNIL, spells this out in its framework on cookies and other trackers: consent must be freely given and explicit. The result: a share of your real audience never shows up in your reports. Depending on the sector, this refusal rate alone explains why your tool often underestimates your real traffic by 10% to 30%.
Sampling plays a part too. On large volumes, some tools compute on an extract, then extrapolate. Each platform defines a session its own way: 30 minutes here, a different threshold elsewhere. I never compare two tools against each other. Pick one as your single reference and track its trend over time. That's the only honest comparison.
The 5 statistics to track first (and which to ignore)
Five statistics are enough to steer a website: number of visitors, traffic sources, page views, session duration and conversion rate. The rest feeds reports, not decisions. I've seen too many owners drowning under 15 metrics they couldn't tie to a concrete action.
To sort them, I use a simple grid: the decision versus vanity matrix. A decision metric changes what you'll do tomorrow. A vanity metric is pleasant to read but guides no action. The number of "likes" on a page? Vanity. Conversion rate by source? Decision.
| Statistic | Type | What it tells you | Decision it triggers |
|---|---|---|---|
| Visitors / sessions | Decision | Audience volume and loyalty | Invest more or less in acquisition |
| Traffic sources | Decision | Where your customers come from | Double down on the channel that converts |
| Page views per session | Context | Depth of browsing | Improve internal linking |
| Session duration | Context | Interest in the content | Rework a page people leave too fast |
| Conversion rate | Decision | The site's real effectiveness | Fix the leaking funnel |
| Bounce rate alone | Vanity | Almost nothing without context | None, until it's cross-referenced |
The three decision metrics deserve your attention every month. The two context metrics are worth checking when a decision figure drops, to understand why. Bounce rate on its own has no place in your dashboard.
What is a website's audience?
A website's audience is all the people who visit it over a given period, measured mainly by the number of unique visitors and sessions. It's the first figure to know, and rarely the most useful on its own.
A local service brochure site often aims for a few hundred to a few thousand visitors a month. A niche e-commerce site can run around 5,000 to 20,000. These orders of magnitude are worthless in absolute terms: 300 ultra-qualified visitors who request a quote beat 10,000 curious ones who leave right away. An audience is judged on its quality, not its raw volume.
How many statistics do you really need to watch?
Five to six statistics are enough for a small business. Beyond that, the time spent reading outweighs the time saved deciding. I apply a strict rule: if a metric triggers no action after three months, it comes off the dashboard.
This principle matches the logic of the 10 marketing KPIs to track when you run a small business: better a few numbers well tied to your goals than lots of decorative charts. Simplicity is a management choice, not a technical limit.
Where your traffic comes from: decoding sources
Traffic sources split your visitors by the channel they arrive through: Google search, direct link, social media, referring sites or advertising. It's the most strategic information of all. It tells you where to put your energy and your budget.
Google Analytics 4 automatically classifies traffic into broad channels. Organic traffic comes from natural search. Direct traffic corresponds to people who type your address or use a bookmark. Referral comes from another site. Social comes from social networks. Paid comes from your campaigns. Organic search accounts for a major share of web traffic, as shown in our SEO statistics for small businesses in France.
Direct, organic, referral, social traffic: what do they mean?
Each channel tells a different intent. Organic traffic shows Google judges you relevant for certain queries. Referral shows other sites cite you. Social measures your reach on the networks. Direct, in theory, reflects awareness.
In practice, this classification is more fragile than it looks. An untagged email, a link shared on WhatsApp, a mobile app: all of it frequently ends up filed under "direct" for lack of origin information. That's where the most common trap hides.
Why direct traffic is often overestimated
High direct traffic is not a sign of awareness. Most often it's a symptom of poor tracking. Contrary to what you read everywhere, you should try to reduce it, not celebrate it.
Direct traffic often exceeds 30% of sessions on small business sites. Yet a large part of it is nothing like "direct". These are clicks from emails without tracking tags, private messaging apps, or apps that don't pass along the source. This phenomenon is called dark social, invisible social traffic. When Analytics doesn't know where someone came from, it files them as direct by default.
The consequence is twofold. You overestimate your awareness. You also underestimate your truly high-performing channels, email or digital word of mouth. The fix comes down to one practice: systematically tagging your outbound links with UTM parameters (labels added to the URL that indicate the origin). A link in a newsletter, a campaign, a post: each must carry its label. Without that, you're steering blind.
When I see direct traffic above 35% on a site that emails regularly, I know before I even dig that the tracking is broken. Fixing it often makes a previously invisible email channel "appear", and that changes budget decisions.
Reading duration, page views and bounce rate correctly
Session duration, page views and bounce rate mean nothing in absolute terms. They're always read by page type and by source. The same figure can be excellent on one page and disastrous on another.
Take a blog post. A visitor reads everything, finds their answer, leaves without clicking elsewhere: they generate a "bounce" and a short session. A failure? No. They got what they were after. The same behavior on an e-commerce product page signals a problem.
Here are the normal ranges I observe by site type. Benchmarks, not absolute truths.
| Site type | Average session duration | Pages per session | Typical bounce rate |
|---|---|---|---|
| Tradesperson / local service brochure site | 1 to 2 min | 2 to 3 | 40 to 60% |
| E-commerce | 2 to 4 min | 3 to 6 | 30 to 50% |
| Blog / media | 1 to 3 min | 1 to 2 | 60 to 80% |
| B2B services site | 2 to 3 min | 3 to 5 | 40 to 55% |
Is a high bounce rate necessarily bad?
No. It depends entirely on the page in question and the visitor's intent. On a blog, a 75% bounce is normal. On an e-commerce homepage, it's a warning sign.
Never judge a bounce rate without cross-referencing it with the landing page and the source. A high bounce on paid traffic is expensive, because you're paying for clicks that leave. The same bounce on organic traffic to informational content is often healthy. This cross-reading is part of a proper website UX analysis step by step.
What visit duration is normal for a site?
Average session duration hovers around 2 to 3 minutes on a brochure site, with wide variation depending on the type of content. Below one minute on a page meant to hold attention, look for a friction point.
Watch out for a technical trap. Since Google Analytics 4, the concept has changed: we now talk about "engagement time" rather than classic time on page. This duration only counts the time the tab is actually active. Your figures may look lower than before without anything getting worse. It's a change in how it's measured, not in your performance.
Linking traffic to results: conversions and ROI
Conversion rate is the only statistic that links your traffic to your results. It measures the share of visitors who complete the intended action: purchase, quote, getting in touch. The referee. Everything else is just a means to get there.
The average e-commerce conversion rate sits between 1.5% and 3% depending on the sector, with the top stores exceeding 3.2%, as detailed in our e-commerce conversion rate statistics in France. A brochure site is judged on its contact or quote-request rate, often far higher on qualified traffic.
How do I know if my traffic makes money?
Your traffic makes money when you can assign a value to each conversion and compare it with the cost of the channel that generated it. Without that dollar figure, you're not steering. You're guessing.
The method comes in three steps. First, define your conversions in Google Analytics 4: a purchase, a submitted form, a triggered call. Then, assign an average value to each, even a rough one (a quote might carry a 15% chance of signing a €2,000 contract, so €300 of expected value). Finally, cross that value with the traffic source to know which channel truly fills the till.
What I see most often illustrates the payoff well. An online store selling seasonal products was tracking fifteen tables without ever linking its traffic to actual sales. By refocusing everything on five metrics and assigning a value to each conversion, the team discovered that its email channel, filed under direct traffic, generated half of its revenue. They reallocated their ad budget toward email and reworked two leaking pages. The conversion rate rose clearly the following quarter, without a single euro of extra traffic. This logic complements our method to measure SEO ROI in a small business.
How to find out a website's traffic: the tools
To find out your own site's traffic, Google Analytics 4 and Search Console are enough, both free and covering an SMB's core needs. Paid tools only become useful for watching the competition.
Google Analytics 4 measures behavior on your site: who comes, from where, what they do. Search Console shows how Google sees you in search results: queries, positions, clicks. The two complement each other. Matomo is a privacy-friendly French alternative, worth a look if you want to host your data yourself or avoid consent banners in certain setups.
How do you find out a website's traffic?
Install Google Analytics 4 via Google Tag Manager or your CMS, check that collection works, then connect Search Console. In under an hour of setup, you see your real traffic. Our guide to getting started with GA4 as a beginner details every step.
Do you need to pay for a tool as a small business?
No. The Google Analytics 4 and Search Console pairing covers volume, sources, behavior and positions for free. Paying is only justified for two specific needs: closely tracking your keyword positions, or analyzing competitors.
| Tool | Indicative cost | For which need |
|---|---|---|
| Google Analytics 4 | Free | Traffic and behavior on your site |
| Google Search Console | Free | Queries and positions in Google |
| Matomo | Free (self-hosted) to paid | Privacy-friendly alternative |
| Semrush / Similarweb | €100 to €140/month and up | Estimate competitors' traffic |
| SE Ranking | €40 to €100/month | Position tracking on a controlled budget |
For position tracking, pairing the free Search Console with an affordable tracker is enough in most cases, as shown in our comparison of Google rank tracking tools for small businesses.
How do you estimate a competitor's traffic?
You estimate a competitor's traffic with tools like Semrush or Similarweb, except these are only estimates, never exact figures. These platforms have no access to the real data: they extrapolate from panels and models. Use them to compare orders of magnitude or spot a rival's keywords, not for a precise audit. France Num makes the same point in its advice on improving a website's effectiveness: tracking your own traffic matters more than competitor monitoring.
Building a simple dashboard and reading it every month
A useful dashboard fits on one screen. Five to six metrics. Read in 20 minutes a month. Beyond that, it's no longer a steering tool, it's a chore people abandon.
My monthly routine comes down to five questions. Ask them in order, every first Monday of the month. Twenty minutes, no more.
- Is my traffic going up or down? Compare sessions to the previous month and to the same month last year, to neutralize seasonality.
- Which channel brings me the most conversions? Look at conversions by source, not sessions by source. That's different.
- Is a page slipping? Spot the most-visited page whose duration or conversion has dropped.
- Is my overall conversion rate moving? A clear swing signals a technical friction or a change in audience.
- What to do this month? A single priority action, drawn from the four previous questions.
This framework turns a mute dashboard into an action plan. The concrete build of this tracking is detailed in our method to create a useful Google Analytics dashboard. The classic mistake: wanting to measure everything from the start. Begin with these five questions, add more only when a gap makes itself felt.
Which metrics belong on a monthly dashboard?
Put five metrics: sessions, traffic sources, conversion rate, top-performing page and conversions by channel. These are the only ones that trigger a decision. Cross-referencing Search Console and Google Analytics 4 by hand takes about thirty minutes a month, or happens continuously if you automate the tracking of your metrics. The real cost here isn't the data. It's the time to gather it every month.
Taking back control of your numbers without burning your evenings
You now know which statistics matter and how to read them. The real obstacle stays the same: finding the time to gather them each month, across Google Analytics 4, Search Console and your campaigns, without losing your evenings to it. That's exactly where owners who grasped the theory get stuck. It's the work we industrialized in Lysible, because no small business owner redoes this cross-referencing by hand every month with any consistency. The data is already there, free. What's missing is a single place where it becomes a decision, not one more monthly chore.
Frequently asked questions
How do you find out a website's traffic?
Install Google Analytics 4 on your site, via your CMS or Google Tag Manager, then connect it to Search Console. These two free tools give you the number of visitors, their sources, the pages viewed and your positions in Google. In under an hour of setup, you see your real traffic. Check that collection works by visiting your site from your phone and watching the real-time report. For your own site, no paid tool is needed: this pairing covers the full needs of a small business.
What's the difference between a visitor and a visit?
A visitor is a unique person, a visit (or session) is a stint on the site. The same person coming back three times in a month counts as a single visitor, but three visits. That's why the number of sessions always exceeds the number of unique visitors. A high ratio between the two signals loyalty or hesitation before buying, common in services and B2B. Understanding this gap keeps you from confusing awareness with volume of visits, two very different realities to interpret.
Is a high bounce rate necessarily bad?
No, a high bounce rate isn't necessarily bad. It all depends on the page and the visitor's intent. On a blog post, a bounce of 70 to 80% is normal: the person read, found their answer and left satisfied. On an e-commerce homepage or a paid landing page, the same figure raises a flag. The rule: always cross-reference the bounce with the landing page and the traffic source. A high bounce on paid traffic is expensive, whereas it's often healthy on organic informational content.
How long does a visitor stay on a site on average?
Average session duration hovers around 2 to 3 minutes on a brochure site, and varies by content type. An e-commerce site often exceeds 3 minutes, a blog stays shorter. Careful: Google Analytics 4 now measures "engagement time", that is, the time the tab is actually active. Your figures may therefore look lower than with the old Analytics, without your performance having dropped. Below one minute on a page meant to hold attention, look for a friction point: slow loading, disappointing content or poor targeting.
Do you need to pay for a tool to analyze your traffic?
No, not to analyze your own site. Google Analytics 4 and Search Console, both free, cover volume, sources, behavior and positions. Paying is only justified in two cases: fine-grained position tracking on strategic keywords, with a tracker at €40 to €100 per month, or competitive analysis via Semrush or Similarweb, which are pricier. For a small business, these tools are monitoring add-ons, never the foundation of your steering. Start free, add a paid tool only when a specific need appears.
How do you estimate a competitor site's traffic?
You estimate a competitor's traffic with Semrush or Similarweb, but these are estimates, not exact figures. These tools have no access to the competitor's real data: they extrapolate from browsing panels and statistical models. Use them to compare orders of magnitude, spot the keywords that earn a rival money, or identify its main traffic sources. Never take them for precise values. For your own site, only your Google Analytics 4 data is 100% reliable, within the limits of consent.
Which metrics should you track first when starting out?
Track five metrics to begin with: the number of sessions, traffic sources, conversion rate, top-performing page and conversions by channel. These five figures trigger concrete decisions, unlike bounce rate alone or raw page-view counts. Look at them once a month, compared to the previous month and the same month last year to neutralize seasonality. Add other metrics only if a real need appears. Too many metrics from the start always leads to abandoning the tracking.
Why do two tools give different numbers for the same site?
Because three factors vary from one tool to another: rejected cookies, data sampling and the definition of a session. Every visitor who declines consent becomes partly invisible to Google Analytics 4, which underestimates real traffic. Some tools compute on a data extract then extrapolate. And each platform defines a session its own way, with different inactivity thresholds. The practical takeaway: never compare two tools against each other. Pick one as your single reference and track its trend over time, that's the only honest comparison.
Isaac SIKORSKI
With Lysible, I want to give businesses back control of their online presence. A website you actually understand is one that brings in real opportunities.
