Targeted use of AI and data

AI and analytics: new opportunities for reputation management

Companies are increasingly making their business decisions based on data. Digital technologies are constantly creating new “sensors” that provide information and can be used for digital processes. For the first time, this enables an informal survey of the entire process from product development to sales, marketing and corporate communications. The self-optimization of corporate processes makes it easier to adapt the management of value creation to changing market, competitive or regulatory conditions.

In addition, there are social trends and changes driven by a wide variety of stakeholder groups that companies are increasingly unable to ignore in order not to jeopardize their business model. This is why companies and organizations today monitor news and discussions around the world around the clock. Traditional analysis and reporting systems support corporate communications in maintaining an overview and separating relevant from irrelevant information. At the same time, digital technologies are changing the traditional tools of corporate communication and are permanently increasing the demands placed on professional corporate communication. In order to maintain a dialog with stakeholders in the future, corporate communication must deal with big data and artificial intelligence technologies as well as new communication platforms in the long term. This will inevitably make corporate communication more data-driven and automated.

This is also the assessment of the European Communications Monitor 2016, which for the first time this year investigated the knowledge, perceived importance and use of big data in corporate communications. More than 2,700 experts in 43 European countries were surveyed for this initial assessment. The results of the survey provide fields of action and implications for corporate communications in dealing with real-time monitoring and data analytics as a basis for developing strategies to safeguard corporate reputation. However, even though three out of four communications professionals in Europe (73%) believe that big data will change their profession, the study shows that big data has been little used in practice to date. Of eleven possible reasons, the lack of analytical skills (49 percent), too little time (45 percent), the lack of technical skills (37 percent), budget reasons (24 percent) and organizational barriers (23 percent) are cited as the biggest obstacles to the use of big data in corporate communications. Added to this are the possibilities of using artificial intelligence in analysis based on big data.

Data-driven communication for better decisions

Basically, big data and analytics are about the structured collection, analysis and interpretation of large volumes of data. The data that can be analyzed for this purpose is growing and changing at breathtaking speed. Corporate communication is therefore challenged to collect very large amounts of data, increasingly in real time, and to interpret it in such a way that a reliable basis for decision-making is created that supports management in evaluating and qualifying business decisions in a targeted and efficient manner. To this end, corporate communications must decide how big data can be used for strategic communication, which requirements must be met in order to use big data successfully and in which areas big data applications should be used. This also changes the process of how communication is planned and managed on the basis of big data. There is also the question of which big data and analytics applications are used and how they are orchestrated. Current process models outline a typical process comprising seven steps: 1. target definition, 2. data generation, 3. data cleansing, 4. data transformation, 5. data analysis, 6. evaluation and 7. preparation of results. Every corporate communications department must adapt and implement this process to its own conditions. This is a challenging and complex task that involves trial and error and feedback loops until the overall process can unfold its effectiveness and reliable results can be obtained. Each of these steps in the process model is associated with specific challenges. As the supposed added value of big data lies in the evaluation of large volumes of data, data analysis plays a special role in the overall process.

The four stages of data analysis

Big data analytics can be divided into four levels of complexity. Descriptive analysis provides information on any variable, for example the number of “likes” or “page views”. The information collected can either be evaluated in real time or viewed retrospectively. Diagnostic analysis goes one step further and enables a deeper analysis of reasons, correlations and patterns. Based on empirical data and using algorithms, the virality of a topic or the career of a message, for example, can be presented together with the reasons for its development. Predictive data analysis goes one step further and enables forward-looking statements to be made. This makes it possible to predict the probability of future events. Several predictors are combined to form a prediction model in order to analyze future probabilities with an acceptable degree of reliability. In predictive analytics, data is collected and transferred into a statistical model to make inferences and predictions. If this process takes place in real time, the model is constantly validated and adjusted if additional data is available to further enrich the model. In practice, predictive data analysis is increasingly being used in research or in the industrial production environment à la Industry 4.0. However, this requires data volumes that are often not available in order to make appropriate predictions using predictive analyses. For this reason, predictive analyses in corporate communications are currently still very rare, even if new applications for specific areas of use will come onto the market in the future. Experts see areas of application primarily in online marketing (cumulative sales data to forecast sales trends more accurately), in issue and stakeholder management (deriving attitudes and behavior in special situations or crises) or in reputation management (dissemination and perception of product and brand messages).

From data to optimal decisions

The supreme discipline is prescriptive analysis. It not only delivers descriptions or predictions, but also provides concrete decision support and recommendations for action derived from it. The prerequisite for this is the linking of different data silos along a communication process, such as between a customer relationship management system, marketing automation systems or for the distribution of content and sentiment or network analyses. Prescriptive analyses are related to descriptive, diagnostic and predictive analyses. While descriptive and diagnostic analyses provide insight into what has happened, and predictive analyses help to model and predict possible future events, prescriptive analyses seek to determine optimal solutions or the best outcome among various available choices based on known parameters. Prescriptive analytics can also suggest decision options to take advantage of a future opportunity or mitigate a future risk and illustrate the consequences of decision options. In practice, prescriptive analytics can continuously and automatically process new data to increase the accuracy of predictions and provide better decision options. Advances in processing speed and the development of complex mathematical algorithms applied to data sets have made prescriptive analytics possible. Specific techniques used in prescriptive analysis include optimization, simulation, game theory and decision analytic methods. Today, prescriptive analysis is more likely to be found in the field of business analytics (BA) than in the field of corporate communications. However, it is highly likely that prescriptive analyses will also be used in corporate communications in the coming years in order to find optimal approaches for special situations and issues. The rapid development of data-based and automated big data and AI applications, which are currently being used primarily to optimize industrial production processes, points the way forward and shows what possibilities corporate communication will have in the near future to sustainably improve value creation in companies.