Hates Google Ajax

20 Reasons Google Analytics Might Incorrectly Measure Your Page Views (PIs)

Page impressions (PIs) are one of the most important units in Internet business. They show you whether a website is doing well or badly compared to others and whether certain changes to the landing pages have a positive or negative effect on traffic. What are the reasons Google Analytics might incorrectly measure their page views (PIs)?

In addition to the page views, there are plenty of other important key figures that define a page: bounce rate, conversion rate, number of visitors, pages per visit, time spent on the page, etc. However, each and every one of these figures is inevitably linked to the page views. So if an error creeps in when counting the number of page views, this has a major impact on all other performance indicators.

The majority of websites today use Google Analytics as a traffic management tool. Although users have been somewhat annoyed about various developments in recent years (one example is the “not set” in many reports and the associated lack of transparency), it still offers the best option on the market to access a wide range of data about the to get to your own website.

Incorrect measurement of page views - what to do?

Sometimes it happens that the pageviews are consistently too low or too high and there is no logical explanation for this. Or suddenly there is a rapid, temporary increase that no analyst understands.

After circumstances that were caused by the company itself, such as a successful advertising campaign, a longer offline period of the website or normal, seasonal fluctuations in the industry, it is worthwhile to go through the following checklist. The causes can be as diverse as they are simple: Perhaps the tag was not implemented correctly by the programmer, someone made a mistake while interpreting the data, or an external factor is causing this misrepresentation? Find out with this checklist!

Is it because of the implementation of the tag?

Use of outdated code

With the many new versions of Google Analytics, the requirements for the day have changed too. The code snippet that had to be installed three years ago, for example, differs significantly from the current one. In many cases the older versions still work, but they often leave a lot to be desired in terms of accuracy. To prevent inaccurate data, it is best to switch to Universal Analytics right away. In this context, you can consider installing the Google Tag Manager right away and having Universal Analytics triggered via the Tag Manager. A brief introduction to both can be found here and here.

Built-in double tag

It may sound banal, but it happens a lot in practice: a new tag was installed correctly, but you forgot to remove the old one. The result: Both tracking work at the same time and generate twice the number of page views.

Parts of the tag are missing or incorrectly changed

If several departments are involved in the implementation of the Google Analytics Tag, it can happen that parts of the original code are suddenly missing. It is therefore important that in the last instance the completeness and correctness of the built-in tag is checked. Otherwise it can happen that the main pages do not appear in the report at all. Even a minimal change, such as the use of typographical quotes that are slightly oblique instead of the normal straight ones, can lead to such problems. You should therefore be very careful when using a “Custom Code”. If you are not sure how to use this self-created code, you can quickly overwrite the complete links on your own website.

Code installed in the wrong place

If it used to be said that the Google Analytics Code can only be integrated at the end of a page, this standard is already completely out of date. Ideally, the tag is implemented in the of the source code. You can find out exactly how the integration works correctly in the following article.

Local hosting of the "analytics.js" archive

Hosting the Google Analytics code locally on your own server means a lot of extra work and is much more error-prone. Every time there is a change from Google, it has to be uploaded to the server. So it would be better if the data is directly with Google and the updates are automatic.

Problems when using Ajax frameworks

One of the peculiarities of the Ajax Framework is that pages are often not reloaded, only overwritten. This can have an impact on the number of page views. Because in the standard setting a page has to be reloaded so that it is counted as a Page Impression. One solution can be to generate virtual page views.

Problems using iFrames

There is a similar problem with incorrect counting with the integration of iFrames. Here, the content of a third website is mirrored on your own. As a result, there are two page views when the pages are loaded: once for your own website and once for the structure of the mirrored page.

Virtual page views used instead of events

User actions such as scrolling on a website or clicking on a web banner should never be defined as virtual page views. This is because these are not actual page views, but events. As such, they should also be defined in Google Analytics.

Doubling through meta refreshes

Pages that contain many advertising banners are often updated automatically. The user has seen a banner and then the page is updated to include the next banner in the same area. This can also lead to double page views.

Did you make a mistake in interpreting the data?

Bad filters

Filters are a great way to take a closer look at data. But they are also very error-prone. Especially those who include or exclude data can also falsify the standard data. Best precaution for this: In the unfiltered view, where all data is available, filters should never be applied!

The lack of personalized filters

The standard view of Google Analytics is very complex, but of course never meets all requirements. It is important to use exactly the right filters and clean up data. It can often happen that not all channels are properly assigned or that the sources are not properly displayed. You should therefore not do without the creation of personalized filters.

Query parameters incorrectly managed

With the help of the "Exclude URL query parameters" function, certain pages can be excluded from an analysis based on defined parameters. In this case, however, it is important to move these pages to the so-called "custom dimension". Because just because they are not relevant for the current evaluation does not mean that they are not needed for later analyzes and could be of interest.

Triggers and tags: poorly defined or not published

When using Google Tag Manager, it is important to consider when a page view should be counted. Is this the case every time a new page is loaded (All Pages), when a certain symbol is clicked, etc.? The trigger gives the tag the information when a page view should be counted. If this is poorly configured, this leads to irrelevant page views. Also important: Tags and triggers must also be published at the end. It doesn't help if these are only created in the system but not used for counting. So be sure to check at the end.

Are there various other, external reasons?

Adulteration by robots & spam

Your own data can be falsified by automatic page views by robots and various other bots. There are no users behind it, but automated programs. In general, a distinction is made between ghost spam and crawler spam. How can you exclude this artificially generated traffic? One possibility is to create a hostname filter for these pages.

Widgets that access Google Analytics data

The Google Analytics ID can also be used for other widgets, which then also track your own website. But here, too, there can be a double measurement. In addition to the new widget, Google Analytics continues to measure traffic using the already integrated code.

Do not use real-time data for evaluations

With Google Analytics you can view your own page in real time: Which visitors are currently on the page? Which side are you on? Where do the guests come from? How long do you stay? This tool can be very useful and give valuable information about the current state of the page. Nevertheless, you should always plan a time interval when calling up data. In some cases it can take up to several hours until the complete information is listed in the account. You can be on the safe side if you only look at the data from the previous day on the morning of the next day. Then all the data should already appear in full.

JavaScript and cookies turned off

Some users do not allow their computer to allow JavaScript or cookies because they fear security risks. However, Google Analytics needs these functions in order to work correctly. Only a fraction of Internet users currently use such programs.

Ad blocker and script blocker

Many Internet users today feel that the many advertising messages and banners are too much patronized. Ad blockers are a way to protect yourself from annoying advertising content. Although the proportion of their users is still relatively low, this will definitely increase in the future. Certain blockers are even set so aggressively that they block tracking scripts and the Google Tag Manager.

Data sent incorrectly

Someone falsely sends traffic data to your Google Analytics account, or vice versa, are you sending data from your pages to a third-party account? Various host names can be excluded to prevent such mishaps.

Note on sampling

Sampling is a term used to describe the creation of personalized segments. That means you want to break down the standard data differently in order to get the desired reports. It is important to note the following: When sampling, there may be slight deviations in the data.

So those were 20 reasons why Google Analytics measures the page views incorrectly. This checklist is definitely not complete.

Do you know any other reasons that could be the cause? We look forward to continuing the checklist by leaving us a comment below this article.

Get support in web analysis now