We determine which plan you require based on the amount of traffic to your site. Specifically, the number of “visits” to your site in a month. We define a visit as:
A unique IP address in a 24-hour period.
Essentially, we count the number of unique IP addresses that visit a site in a day. This is the number of visits for that day. We then add up all the daily values for the entire month, and that’s how we arrive at the number of visits for your site. Before we do this, we do our best to remove IP addresses from known bots, spammers, and attackers.
We feel that this is the most fair and transparent way to measure traffic. There are many ways to do it, each with their limitations. We feel it’s important to be clear about how we measure usage and not use ambiguous terms.
How many page-views does this equate to?
Because we only count each visit, it means that there’s no difference between someone who visits your site and views one page, or someone that views a hundred. Both cases count as “one visit” to your Flywheel site. So, the number of page-views could be many times the number of visits, depending on how many pages your visitors view each time they visit your site.
Why not measure using page-views, visitors, or some other metric?
Like we mentioned, there are a ton of ways to measure traffic. Some hosting companies use page-views, some use visitors, and some just make things up. Visits is an easy metric for us to measure on the server side, without having to inject any code into your site, which might slow it down.
Can I view a report of my usage?
Only upon request. You can view information about the number of visits on your site, CDN and disk usage in a portal dashboard. At the moment (2020) we do not give access to this dashboard to everyone, but this will be a standard feature soon available.
Wait—my visitor count on Websetgo doesn’t match Google’s. Why is that?
There are several reasons for this, but the short answer is: we’re counting different things and using different methods. We both have our specific algorithms, and we’re both doing the best we can to be as accurate as possible, but we may filter (or more likely, notice) things Google doesn’t, and vice versa.
Not that one is right and one is wrong, necessarily; we just have two different methods. We continually improve our tracking by monitoring traffic to spot new bot types and remove them.
To get more technical: a big reason for the disparity is that Google uses JavaScript-based tracking. That means that Google counts page loads in a JavaScript-enabled browser. Google may not track a visit if the whole page (or at least enough of it to get to the analytics script) doesn’t load, if JavaScript is disabled, or if the request is not loading in a browser window.
Another important note: Google can’t track any visitor who uses an ad-blocker extension or a browser that blocks trackers (like Brave, Firefox Focus, or certain versions of Safari). By most estimates, at least 25-30% of web users implement an ad blocker.
For these reasons, we’re sometimes counting things differently than Google, and will often detect things that they can’t. That said, we do our best to filter out requests we do not define as a “visit” to your site.
Okay, so what do you remove from your visit tracking?
There are two types of visits we remove from visit tracking.
First, we remove any traffic from visitors whose user agent doesn’t look like a web browser. If you’re not familiar, a user agent is a self-reported piece of data included in the web request that identifies the “agent” the request is using to access the site – like which browser is being used. If it doesn’t look like a browser is hitting your site, we conclude it is not a site visit to count against your plan limits.
Second, we evaluate the remaining pool of requests for any that appear to be bots or crawlers. We use an intelligent algorithm for checking for bots and crawlers instead of relying on potentially outdated bot lists.