Digital analysis: Why more is not always better, or 3 strategies to get more out of your data

Using data-based marketing to become more customer-oriented, to organise marketing activities more purposefully: that is the vision. Many decision-makers, however, face a huge amount of web analytics data and are not able to process it systematically so that the collected information could support marketing decisions. This article shows where the challenges lie and what strategies you can use to get more out of web analytics data.

Web analytics, today more commonly referred to as digital analytics, has become an established discipline in many companies and is a regular item on the agenda of marketing meetings. Marketing decision-makers have realised that it is a valuable instrument for controlling their activities in digital channels. However, if we have a look at the entire process from data collection and data storage to data analysis, we find that many companies in the recent years have focused on the first two stages. The amount of data collected is huge: (2) the number of visitors, clicks, bounce rate etc.

The difficulty is to extract the relevant information from this flood of data and use it to make informed decisions. John Naisbitt, a US trend researcher, has thus described the challenge: “We are drowning in information but starved for knowledge.” A study by etracker (3) shows that many businesses use the collected analytics data for internal reporting only. Only one third of the respondents use the data systematically to generate insights for optimising the marketing efforts or use web analytics as an early warning system.

There are many factors hindering or preventing a systematic and structured data analysis:

  • System- and channel-oriented data collection and processing instead of a holistic view of the customer journey.
  • Processing the data using the “one view fits all” approach instead of views tailored to individual stakeholders (analysts, people responsible for the web presence, CMOs, people responsible for campaigns etc.).
  • Isolated presentation of data instead of presentation of logical and/or chronological relations between the data.
  • Static reports instead of dynamic views which make it possible to drill down on detailed data to identify causes and patterns.
  • Juxtaposing a lot of data without aggregating it to strategic key performance indicators and without specifying its relevance for actions.
  • Reporting as a past-oriented accountability mechanism instead of a future-oriented optimisation tool.

3 strategies to use data as support for decision-making

The following three strategies will help you get more out of web analytics. Not by gathering more data but by forwarding the relevant data in the right form, with the right context information, to the right place:

1. Process the data individually:

Web analytics is used more and more frequently by people other than analysts. The circle of stakeholders has grown in the recent years: marketing decision-makers, sales managers, product managers, all of them use web analytics findings in their work. Marketing managers wonder how much traffic and turnover has been generated by a campaign. A content marketer wants to know how the contents appeal to the visitors. A sales manager is curious to what extent the website serves as a lead generator. A product manager wants to know what products with what features are in the highest demand. Certainly, the information needs of these various stakeholder groups vary significantly. The data has to be processed individually to address the individual persons directly. Consider each affected person separately and identify their information needs to define a dashboard that is best suited to their individual situation.

2. Analyse the data holistically to identify patterns and dependencies:

Isolated data, such as the number of visitors or clicks, is generally not useful for explaining specific phenomena. Only when data relationships are considered, not only with respect to time, but also the relations with marketing activities, along the customer journey, or with external factors, will it be possible to recognise patterns and define the appropriate course of action.

3. Visualise the data:

Mere figures, scattered across a huge Excel table, are not enough to recognise patterns or correlations. They may even lead you to false conclusions. With visualisation, the data suddenly starts to tell stories. Visualisation supports information processing. It can be used to present complex issues and show correlations. In a TED talk, David McCandless described the benefit of visualisation as follows: „Visualizing information, so you can see the patterns and connections that matter.“ (4) You can use the following questions to help you prepare relevant visualisations:

  • Who is the recipient?
  • What decisions does the recipient want to back with the information?
  • What information does he need to make these decisions?
  • In what form does he read and interpret the information?

(1) „Information: the currency of the digital age“ Speech on the occasion of Oracle OpenWorld in San Francisco 2004
(2) When collecting data, it is absolutely necessary to follow the data protection guidelines. It is also recommended to always offer the user an added value in return.

Data visualisation – the tool for easy and quick knowledge acquisition

Our first article concerning digital analysis revealed the challenges faced by companies in today’s world of data. Companies often have enormous quantities of data at their disposal but fail to take advantage of them. This is because many entities focus on gathering and storing data, while the aim of digital analysis – which is to enable deriving data-supported recommendations and optimisations of actions – is frequently not pursued consistently. Data should support the decision-making process, and to this end there must be a possibility to analyse and interpret them comprehensively.