Being a Web Analyst basically means bringing order to chaos. Data, understood as a mass of brute and random numbers, can be useless, incoherent, the proverbial block of raw clay in which it seems impossible to guess the form of the work of art that it will yield… Our job is just this: to give a structure and perspective, to transform this primordial broth into an organised ecosystem, from which to obtain actionable insights, i.e., precise logical inferences useful for decision makers to make the right decisions within a valid digital marketing strategy.
The primordial broth of data
We are hearing increasingly about data-driven companies, big data, and insights. The strategic importance of data and their analysis.
The truth is that we are continuing to swim in a sort of primordial broth made of numbers, strings, and tables; a broth with huge potential, waiting for the spark that can bring to light the insights, the analysis.
One of the most important steps an analyst can take, contrary to what one might think, is to reduce the amount of information in favour of quality. The risk of losing sight of the general picture by collecting and processing data of little use is high, and often reduces web analytics to a tool useful only to justify actions already taken, right or wrong they may be, while our aim is to provide evidence-based information to those who then make decisions and formulate strategies.
The setup of an effective measurement system cannot be separated from adequate knowledge of the customer, website/app and the tool used.
Overview: know your customer
Knowing the customer and how his business takes shape in its online and offline components helps us to create the context necessary to identify the micro- and macro-objectives that will then guide the setup of the web analytics tools.
At the same time, correct communication with the customer allows us to save time and be more accurate during the data analysis phase; the activation of online and offline sponsorship campaigns is a typical example of the type of information that any web analyst would like to have a little in advance. Being aware of the marketing activities undertaken by a customer becomes essential to ensure that the platform is then able to collect clean and consistent data from the very beginning.
Understanding the more technical aspects and operating logic of the systems used in data collection and processing not only represents a crucial strategic advantage but is in fact a necessary step if you want to move from a basic measurement system to an advanced one, from a report to an analysis. An in-depth knowledge of tools such as Google Analytics and Google Tag Manager allows a web analyst to use them creatively and effectively.
From this point of view, the organization of work environments becomes fundamental to guarantee consistency in tracking and scalability on growing projects.
Another key point to be addressed in the design phase is data overflow, that is to say, data in excess. Better few but good data.
The enthusiasm and our passion playing “detective” drive us to want to trace everything happening on the website: clicks on buttons, on menu items, on interactive elements on the page, and even tabs, toggles, arrows, and sliders.
Every time we add an element to the tracking we need to be sure that the information linked to it is really important, whether it is a health and performance index or a piece of key information within an evolutionary picture of the website or app. How do we like data? Better few, clean and immediately.
In an industry like web analytics that moves at a high speed and revolves around a changing object like a website, the ability to look beyond the needs of the moment is a major advantage.
Concentrating resources on tracking important elements gives us time to plan activities that go beyond the everyday routine, to define new metrics and more generally to formulate hypotheses and collect data that can confirm or refute them.
A structured and multidisciplinary approach to web analytics allows us to keep in mind the context in which we are moving, obtain cleaner data and consequently produce quality insights.
The setup of the measurement system
Although it may seem trivial, setting up work tools correctly is the most important step in getting clean data.
If it is true that all web analytics projects share a logic, that is, one linked to the collection of basic elements such as pageviews, it is also true that each project needs an advanced setup that is able to take into account the technical and strategic variables linked to the technologies used in web development (e.g., JavaScript frameworks) and those variables linked to the type of business (editorial projects, e-commerce, showcase websites, targets).
The same metrics in different contexts takes on a different importance: for this reason, it is essential to identify and prepare indicators to support the basic ones provided by tools such as Google Analytics.
From simple reports to true analysis: often everything starts from an anomaly. The observation capacity of the web analyst is everything.
The analysis process: from data to insights
Once you’ve created a solid foundation through the technical setup, the crucial moment finally comes: how do you extract strategic information from the data “broth“?
If it is true that the dashboards and reports of the main web analytics tools are full of information and “ready-to-use” data, it is also true that the difference between a report and an analysis is the result of the professional’s ability to find relationships, identify flows and study deviations.
Sometimes the analysis process originates from a detail, an anomaly, like a trend reversal or an unexpected change in a metric; in cases like this, a web analyst focuses on looking for relationships between numbers that can justify the change. This is probably the most frequent case and, all in all, the easiest to manage, as the whole process can be based on a set of data already collected, usually sufficient to create a context and reach a conclusion.
The most complex analyses, however, start from the formulation of a hypothesis and then take shape through a real investigation, which through specific measurement and comparison tools is able to identify patterns that can refute or confirm the initial hypothesis.
It is clear that this is not a simple and straightforward process; the ability to define a method, an approach, and interpret it on a case-by-case basis, is just as necessary as the technical preparation and, indeed, it is a factor whose importance increases together with the degree of complexity of the projects.
