Data-driven marketing is an approach to marketing that starts from data and then arrives at the strategy to be implemented. Through a correct collection and interpretation of data analytics, brands can predict what their customers or potential ones will buy based on their purchasing behavior and their interests. In this way, they are able to send personalized offers, appropriate purchase suggestions and customized campaigns.
The data collected makes it possible to anticipate the needs and desires of buyers, allowing the company to do one to one marketing.
Segment the market and profile users
Marketing starts with segmentation and this, if very detailed, is difficult to put into practice. Market segmentation occurs on the basis of:
- geographic, the market is divided by countries, regions, cities and localities
- demographic, which takes into account: age, sex, occupation and socio-economic class
The geographical and demographic segmentation go from top to bottom and are mostly static, in the sense that a consumer is pigeonholed into a single segment for all products. However, its decision-making process varies according to the product category of the product and what stage of Funnel Marketing it is at. Therefore, marketers prefer a bottom-up approach: instead of disaggregating the market, they group consumers with similar preferences and behaviors; each consumer falls into a segment.
For this approach, other types of segmentation are used, such as:
- psychographic, thanks to which buyers are classified according to: values, interests and personal motivations
- behavioral, the distribution takes place taking into account past purchasing behavior: frequency, amount, types of products purchased at certain times of the year, etc.
Finally, there are the most popular approaches to segmentation today, those that refer to the first-party data, already in the possession of companies. An e-commerce, for example, will have a considerable amount of information about its users, which mix all the previous ones (geography, demographics, spending capacity, purchase frequency, etc.), which must be made homogeneous within grouped and coherent audiences, for example through the RFM approach.
The representation of a segment of consumers obtained thanks to the above techniques is called a persona.
What types of data to consider
Segmenting and profiling users has always been fundamental and thanks to Big Data it is possible to collect new types of data useful for carrying out micro-segmentations. The data comes not only from CRM and market surveys, but also from the media, social networks, physical stores and the IoT (Internet of Things).
Furthermore, they can be classified into:
- Hard Data, which are very personal and therefore hardly subject to change
- Soft Data, which tend to change over time. Soft Data can in turn be classified into: anonymous data, own or from third parties, of persons identifiable only by IP and known data, also own or from suppliers, of persons recognized by name, email, telephone number or other , because in the past they have filled out a form voluntarily providing their personal information
With Big Data, brands can discover ideas for new and personalized products or services, the price to be applied, the most suitable distribution strategies, the creation of content and the selection of media. In addition, they are useful for push marketing, after-sales assistance and customer retention.
Data-driven Marketing: criticality
The difficulty of data-driven marketing lies in integrating all the information obtained with the aim of exploiting it to profile the individual consumer and implement dynamic strategies, obviously with an eye to privacy.
There is the possibility that the results are unsuccessful and the causes may be due to various aspects, such as:
- Focus too much on the choice of tools and less on the marketing project: it is the IT infrastructure that must follow the marketing strategy and not vice versa
- Thinking that big data can replace traditional searches such as usability tests: the two types of data must not be mutually exclusive but complement each other
- Believing that automation can replace the critical observation capacity of the human being: the supervision of an experienced marketer is essential
In addition to this, it is good that the objectives are established upstream and clearly, so that the data-driven approach generates measurable results. The biggest challenge lies in integrating external and internal data and tracing it back to the individual customer, which is not always possible for privacy reasons.