Here are four reasons you may want to do this:
- You’re doing “competitive research” (i.e. trying to figure out how competitive a niche is before launching a website in said niche);
- You’re interested in advertising on a website;
- You’re interested in buying a website;
- You’re looking to write guest posts for sites that actually get traffic.
In general, there are three simple methods for uncovering how much traffic a website gets, which are:
- Use traffic estimation tools;
- Check their advertising page;
- Ask them.
I’ll talk about each of these methods in detail in this post.
I’ll also be sharing my research into four of the most well-known traffic estimation tools, where I’ll be looking into: the metrics they show; how they obtain their data; and how reliable they are (hint: we did a couple of experiments for this!)
Let’s get started!
1. Using Traffic Estimation Tools
Traffic estimation tools can be broken down into these two categories:
- Total traffic estimation tools (i.e. those capable of estimating the overall traffic of a website);
- Organic traffic estimation tools (i.e those capable of estimating only organic traffic)
To ensure a fair comparison, let’s look at both groups separately.
Part 1 — Overall Traffic Estimation Tools
These tools report overall traffic, including: direct; organic (i.e. “search”); social; and referral traffic.
Clickstream data (see definition), and a few other sources, are used to provide this estimate.
Here are the three tools that fall into this category:
Alexa—now an Amazon-owned company—was very popular back in the day, so we felt we should include it on our list.
It’s a pretty simple tool; you enter a domain and it’ll show you a bunch of stats for a given website.
It is, however, extremely unreliable today (keep reading!).
Here are the traffic stats Alexa claims to show:
- Estimated “unique visitors”, “visits”, and “pageviews” (note: these are currently only reported for a handful of countries — however, during our testing, the tool only ever showed us US-traffic stats);
- Percentage of traffic by country (i.e. a geographic breakdown of the websites visitors);
- Engagement metrics (e.g. bounce rate, time on site, average page views, etc);
- Percentage of traffic visiting the site from a search engine (note: it also shows some of the top keywords bringing traffic to the site)
But where is Alexa obtaining this traffic data from?
Here’s what Alexa says:
Alexa’s traffic estimates and ranks are based on the browsing behavior of people in our global data panel which is a sample of all internet users using one of over 25,000 different browser extensions. […] We also gather much of our traffic data from direct sources, including sites that have chosen to install the Alexa script and certify their metrics.
In other words, it seems they’re predominantly relying on clickstream data from their own pool of internet users (i.e. those with at least one of the aforementioned 25k browser extensions installed, along with webmasters who have the Alexa script installed).
Alexa states that their traffic panel is “based on millions of people using over 25,000 different browser extensions.” However they offer no insight into how many million we’re talking about here.
Even if we assume 50 million, it’s still a tiny percentage (roughly 1.4%) of the 3.6+ billion internet users worldwide.
This may be the reason Alexa isn’t reliable. In our experiment, it only returned stats for 30% of the websites we tested.
Similarweb shows a bunch of different traffic-related stats, including:
- Total visits—this is the sum of all visits (non unique) to the analyzed domain, or industry within the time period analyzed (source);
- Pages per visit;
- Average visit duration;
- Bounce rate
It also shows you a graph with the previous 6-months estimated traffic stats, which is nice.
- Traffic by countries (desktop only);
- Traffic sources (e.g. direct, referrals, search, etc);
- Top referring sites (really useful if you’re link-building or looking for guest post opportunities that will actually bring referral traffic!);
- Top 5 organic keywords;
In terms of how these stats are derived, Similarweb does appear to rely on a number of trustworthy sources. Here’s what they say:
Our data comes from 4 main sources:
- A panel of monitored devices, currently the largest in the industry;
- Local internet service providers (ISPs) located in many different countries;
- Our web crawlers that scan every public website to create a highly accurate map of the digital world;
- Hundreds of thousands of direct measurement sources from websites and apps that are connected to us directly.
SEMRush is perhaps more well-known for their organic traffic estimation statistics (more on that later!) but earlier this year, they introduced their “traffic analytics tool”:
Here are the main stats the tool shows:
- Unique visitors;
- Pages per visit (again, you can calculate pageviews manually by multiplying “pages per visit” by “visits”);
- Average visit duration
It can also show a breakdown of traffic by country, medium (e.g. referral/organic/etc).
Right now, the tool currently only shows desktop traffic—mobile traffic is not included.
With Statista recently reporting that almost 50% of all web traffic comes from mobile devices, chances are that SEMRush is drastically underreporting traffic numbers for most websites.
Let’s compare the data to see how these 3 tools stack up [experiment]
In order to see how these three tools stack up against one another, we decided to run a (very small) experiment.
Here’s what we did:
- We used a sample of 37 websites that listed their true traffic statistics on advertising pages;
- We pulled the “total traffic” numbers from each of the aforementioned tools (i.e. Alexa, SimilarWeb, and SEMrush);
- We compared the results.
Like I said previously, Alexa only seems to show US-traffic whereas the other two tools report global traffic.
Alexa also only kicked back estimates for 30% of the sites we checked.
We, therefore, decided to eliminate it from any further comparison.
Here’s how the other two tools stacked up:
Here are a few key takeaways from this data (in plain English):
- Similarweb and SEMRush were pretty much on par with each other correlation-wise;
- SEMRush underestimated unique visitors by roughly 27% on average, yet overestimated “pageviews” by almost 150%;
- Similarweb overestimated unique visitors by almost 300% and overestimated “pageviews” by almost 200%;
IMPORTANT NOTE: The sample size used for this study was very small, thus making it somewhat difficult to reach a definitive conclusion based on this data. We recommend that you try each of the three tools on your own websites and decide which one seems more accurate to you.
Part 2 — Organic Traffic Estimation Tools
Let’s take a look at them in more detail.
Ahrefs can show estimated organic traffic stats for (almost) any website.
We show global organic traffic by default (i.e. under the “overview”) but you can also see traffic segmented by country under the “organic search” tab.
These numbers are calculated by looking at the keywords for which a site ranks. We take into account the monthly search volumes and ranking position for each keyword.
It’s also worth noting that this is a “live” number—it’s calculated for the last month, so numbers may vary slightly from day to day. This number includes both desktop and mobile traffic.
In terms of where our data comes from, we essentially reverse engineer Google rankings.
We use our own keyword database, containing approx. 402M keywords (114M of which are keywords searched from within the US), to calculate global search traffic estimates. This database is updated monthly, including the search volume estimates (check out a few fun search stats from March here)
SEMRush reports global search traffic within their “traffic analysis” tool.
You can also segment by region (i.e. country).
It should be noted that, unlike Ahrefs, the numbers shown by SEMRush are historical by default.
This means that rather than showing a “live” estimate (i.e. for the past month), SEMRush will show figures for the most recent full month; this is usually the previous month’s stats.
SEMRush, like Ahrefs, reverse engineer Google rankings to create their own database of keywords. Again, I failed to find find any information about the size of this database (i.e. the number of keywords they monitor) on their website.
Let’s compare the data to see how these 2 tools stack up [experiment]
It’s time for another quick experiment. Here’s what we did:
- We took a sample of 112 websites from Flippa (all of which showed verified global organic search traffic numbers);
- We pulled organic traffic numbers from both Ahrefs and SEMRush for all 112 websites;
- We compared them.
Here are the results:
We ran both Spearman and Pearson correlations for the data—along with a linear proximity test—and found the following to be true:
- Both Ahrefs and SEMRush were pretty much equally consistent in their traffic estimations;
- Ahrefs underestimated traffic by 60% (on average), whereas SEMRush underestimated traffic by more than 87%.
IMPORTANT NOTE: Ahrefs shows total traffic estimation (desktop + mobile) whereas SEMrush only shows Desktop traffic (with no sum of both available). It’s very likely that SEMRush is lacking on the proximity front primarily because they don’t include mobile traffic in their estimations.
It should also be noted that a sample size of 112 websites is still very small and thus, these should not be viewed as definitive. If we took a sample size of 1000+ websites, the numbers could be different.
If you have a few websites of your own, we would recommend testing both Ahrefs and SEMRush your own numbers. We would love to see the results. We’ll even help with pulling data for a large list of websites if required…just shoot us an email!
But anyway, traffic estimation tools aside, how else could you estimate a website’s traffic?
2. Check their advertising page
Most bloggers receive so many “I’m interested in advertising on your site. How much traffic does it get?” emails that it generally makes sense to simply put this information on a publicly accessible advertising page.
For example, check out the GeekWire.com’s advertising page.
It states their monthly unique visitors, pageviews, and some other useful information (e.g. social stats). It’s not just the big sites that do this. I’ve seen a lot of individual bloggers doing this too.
But, this tactic isn’t foolproof. There are a few potential problems, including:
- Not all bloggers publish these numbers on their advertising page (see the next tip for a solution to this!);
- These numbers aren’t always up-to-date; sometimes bloggers add these numbers when they initially launch their website but never update them; this results in figures that are often months/years out-of-date.
When using this tactic, it’s best to look for blogs that publicly state when these numbers were last updated. Here’s an example from TravelFashionGirl.com:
It clearly states that the figures were updated in March 2017 (i.e. very recently), so these figures can likely be trusted.
If the website hasn’t updated their traffic stats recently—or simply fails to give a “last updated” date—it’s unlikely that they’ll be accurate, so don’t trust them!
3. Ask Them
By far the most accurate (and obvious) method for obtaining a website’s traffic statistics is simply asking the webmaster/blogger.
But most bloggers/webmasters aren’t going to give out their traffic stats to anyone who emails them. So you need to have a genuine reason for asking (e.g. you’re interested in advertising with them).
Here’s an example email that I’ve used in the past when trying to find out traffic statistics:
Most bloggers will be happy to share up-to-date traffic numbers with genuine advertisers; many will even screenshot Google Analytics data as proof.
While this method is typically the most reliable way to decipher traffic stats, it’s not 100% foolproof. Here’s why:
- Some bloggers/webmasters will fabricate traffic stats (in order to get the advertisers money);
- Not everyone has Google Analytics installed. Many will rely on less than accurate analytics platforms (e.g. WordPress plugins) for their data;
- Google Analytics can be easily mis-installed and thus, data isn’t always accurate. If reported traffic sounds too high, try checking their website for duplicate GA code (or ask them what their bounce rate is—a super-low bounce rate is often a good indicator of duplicate GA code)
We recommend looking to traffic estimation tools (as mentioned above) to confirm numbers quoted by webmasters/bloggers. If they’re in the same ballpark, they’re most likely accurate.
BONUS: a few more tips for (roughly) estimating traffic
Estimated traffic stats will usually give you a good starting point, but you can tell a lot about the popularity of a website by manually checking a few things, such as:
- # of comments on their posts (on average);
- # of YouTube video views;
- # of social shares;
- Engagement levels on their fan pages (i.e. Facebook page, Twitter, etc.)
If you take a look at the Ahrefs blog, for example, you’ll see that virtually every blog post has a good number of comments.
Because most readers won’t bother to leave a comment (no matter how great the content happens to be), a blog that receives a consistently high number of comments on their posts will likely be getting a significant amount of traffic.
They also clearly have very active and engaged users.
“Social shares” can be another indication of high traffic. For example, most of the posts on the Ahrefs’ blog receive a ton of social shares. You can easily see that in Site Explorer > “Best by shares” report:
And finally, there’s YouTube.
Looking at our YouTube channel, you’ll see that all of our videos have a decent number of views (most are in the thousands). Again, this confirms that we have a decent amount of traffic.
If a website didn’t have decent traffic—and an engaged audience—these numbers would be a lot lower.
In short, while it is certainly possible to obtain website traffic stats from a variety of tools, the reliability just isn’t there.
Your best bet will always be to ask directly. This is the only way you’ll get 100% accurate stats.
If you use any of the other methods mentioned in this post, keep in mind that this will be a very rough estimate only. Some tools will overestimate while others will underestimate.
If you know of any other (reliable) ways to estimate/find website traffic stats, let us know. We’ll happily add them to the post (if they work!)