List

Web Analytics Frameworks are an essential tool for the increase of acceptance and conversion rates. Nowadays, no firm will be successful without a web analytics strategy or using related tools like A/B testing. In this post, five different web analytics frameworks are evaluated.

Introduction to web analytics

Web Analytics is a significant opportunity to gain knowledge of the data collected on the web. Tools in this area are mainly used to analyze the behavior of visitors on a website (or a webshop). The interests of the analyses serve to:

  • Optimize the website for higher Coverage.
  • Optimize the website to achieve a better Customer Experience.
  • Make the best out of advertising Campaigns.

The easier it is for visitors to find the website and comfortable to use it – the higher is the possibility to convert this traffic into paying customers or returning visitors. The overall goal of the so-called Conversion Rate.

Criteria for web analytics frameworks

In order to compare web analytics frameworks, as well as to review its pros and cons, criteria have to be established. Those criteria can, of course, differ according to the size of the firm, the underlying business purpose, the produced products and services advertised, the audience, etc. Some possible criteria are:

  • The Pricing model (open-source vs. closed-source).
  • Data Collection environment (Client vs. Server).
  • Market Share (incl. Customer Reference).
  • Data Ownership (storage place, access).
  • Use-case suggestions.
  • Analytics capabilities (dashboard, alerting AI).
  • Supporting services (integrations, mobile apps).
  • Partner Ecosystem (data export, service integration).
  • User-friendliness (low to high).
  • Knowledge Barrier (low to high).

The underlying architectures

The data collection environment is based on the underlying architecture. Two different approaches exist according to the data collection process. The server-side collection of data is based on web-server log files and access logs. The client-side collection is based on users’ behavior and preferences. Web analytics requires four different components to be successful:

  • Collection of data from different sources like: website, app, and other internet-connected services.
  • Configuration to customize the collected data and definition of different sets of rules for further processing.
  • Processing of raw data to generate a cleaned dataset, which can be turned into reports that can be interpreted and monitored.
  • Reporting of data, which was collected by visualizing it in different dashboards and Key Performance Indicators (KPI).

Comparison of recent frameworks

In the meantime, the market for web analytics frameworks is of course diverse. However, some systems clearly dominate the market. The following three tools are part of the leaders: Google Analytics, Amazon Web Services, and Adobe Analytics. Below, we compare these three systems, and two alternative tools additionally: StatCounter, and Fathom.

All tools are compared according to the above-mentioned criterion, which are: pricing, webserver, market share, data ownership, use-cases, analytics capabilities, supporting services, partner ecosystem, user-friendliness, and knowledge barrier.

Google Analytics

Google Analytics is one of the most popular platforms for web analytics purposes, especially for traffic analysis. It has a market share of 84.2 % [1]. It offers free reports with information about visitors, traffic sources, e-commerce, and keywords that are bringing the most visitors to the web pages. The related architecture under Google Analytics is broken down into four basic levels [2]:

  • Collection is responsible for collecting user-interaction data.
  • Configuration allows you to manage how the data is processed.
  • Processing operates the user-interaction data with the configuration data.
  • Reporting provides access to all the processed data.

The basic Google Analytics tool is open-source. The data collection is server-side, and the ownership is by the Google Analytics server. The use-cases supported are web analytics, advertising, campaign performance, audience characteristics, audience behavior, site performance, and app performance. The analytics capabilities include dashboards, notifications, reports, and the export of reports. The user can rely on a good support service: support center, online documentation, community support, and an online learning platform. The partner ecosystem is also extensive. Of course the user can use the Google ecosystem, and Power BI via Google Analytics-Connectors is supported. According to my experience, the user-friendliness is medium, same for the knowledge barrier.

Amazon Web Services

The cloud platform Amazon Web Services (AWS) provides plug-and-play tools for operating a web-analytics solution for different web applications. AWS has the highest market share regarding cloud infrastructures. It is 33 %, followed by Microsoft Azure (18 %), and the Google Cloud (8 %) [3].

AWS is not open-source, as users have to pay per use. AWS supports client-side, and server-side data collection. The data ownership is by the AWS infrastructure. The analytics capabilities include streaming analytics, altering, dashboards, reports, artificial intelligence, and machine learning. Analogs to Google, AWS provides a good supporting service. This includes a support center, online documentation, and community support. The partner ecosystem is based on the AWS marketplace. According to my experience, the user-friendliness is low, but this is also because of a high knowledge barrier.

Adobe Analytics

Adobe Analytics is one of the leading web analytics tools. According to the total market share, not even 1 % of websites are using this tool [4]. However, it is heavily used by leading internet retailers.

Adobe does not tell details about the pricing, as this is individual per package. Adobe supports client-side, as well as server-side data collection. The data ownership is in the cloud, it is provided as so-called Software as a Service (Saas). The use-cases supported include web analytics, marketing, attribution, and streaming. The analytics capabilities include dashboards, reports, maps, predictive analytics, and data export. The user can rely on a good community and a support center. The partner ecosystem consists of the Adobe ecosystem and Microsoft Power BI. The user-friendliness is high, and the knowledge barrier is low.

StatCounter

StatCounter is used on 0.9 % of all websites. The user has to pay per page view, and the webserver performs on the client. The data is stored on the StatCounter server, and various web statistics are supported. The analytics capabilities include dashboards, reports, charts, maps, and alert-functions. The support is limited to email and live chat. The partner ecosystem is also limited, to Google Analytics. The user-friendliness can be rated as high, and the knowledge barrier low.

Fathom

Fathom Analytics is a privacy-focused web analytics approach. The user has to pay the tool by a monthly subscription. The webserver performs on the client. About 250 million page views have been analyzed using this tool. The data ownership is in the cloud (SaaS). The use-cases are limited to privacy-focused web analytics. The analytics capabilities include dashboards, reports, and data export. The support is limited to email support and online documentation. The tool is available as WordPress plug-in. The user-friendliness can be rated as high and the knowledge barrier low.

Conclusion

The market for web analytics frameworks is diverse. Nowadays, many tools exist. The comparison above shows that most tools make use of a client-side data collection. Most tools support visualization through dashboards and provide (custom) reports. Generally, a bigger market share of the tool means that a bigger ecosystem is supported. However, always, the underlying use-case should determine the choice of the tool.

References

[1] W3Techs (2020). Market share trends for traffic analysis tools. https://w3techs.com/technologies/history_overview/traffic_analysis

[2] MARKETLYTICS (2020). Google Analytics Architecture Explained for Beginners. https://marketlytics.com/blog/understanding-google-analytics-architecture/

[3] Statista (2020). Amazon ist die Nummer 1 in der Cloud. https://de.statista.com/infografik/20802/weltweiter-marktanteil-von-cloud-infrastruktur-dienstleistern/

[4] W3Techs (2020). Usage statistics and market share of Adobe Analytics for websites. https://w3techs.com/technologies/details/ta-omniture