Why use machine learning in marketing?
Overview of new practices:
The data sources used in customer marketing are diverse. The traces left online by internet users are numerous. Marketers can easily observe consumers through their online shopping, apps, conversations on social media, browsing history or search engine queries.
The ability of marketing departments to collect and process their customers’ data revolutionizes three aspects of business. First, a major change of attitude can be observed: instead of intuition, now one relies on consumer trends, consumer insights and analysis of real behaviors. This is one of the greatest strengths of data: it analyzes the existing data to bring to light new customer behaviors which were previously unsuspected. Once put into perspective, this data becomes information that provides real added value in segmentation, targeting and managing customer relations. For example, a telephone company mapped the profiles of its customers and found out those who were more likely to shift to the competing companies and with this information, it devised strategies to contact and retain them.
Data and consumer contact points
Connecting with consumers in a personalized and ad-hoc manner is a key challenge for advertisers. As a part of their sales strategy, both online and in retail, brands use geolocation-based tools such as geofencing and retargeting to be able to respond immediately and reach their potential customers automatically. Geofencing enables advertisers to send promotional messages and advertisements directly on the mobile phone of a consumer when he is near the shopping center or a concerned shop. This method is yet to be improved by advertisers and retail professionals. Retargeting allows brands to buy web banner ads in real time to track a visitor who has recently visited their website. Though it is an effective strategy, it sometimes leads to a sense of tracking or intrusion among internet users. To optimize the use of data and satisfy consumers at the same time, a few rules should be followed.
Overcoming hurdles in Big Data usage
When an advertiser invests in big data to strengthen his marketing strategy, he must take care of some key points. First, a dialogue with customers is important. It’s important to reassure them and educate them as to how their data will be used. This approach is useful and effective: Studies show that clients are more likely to share more personal information if they feel that they are gaining by this. Personalizing a product or co-creating an offer also encourages data sharing.
Other than customers’ approval, switching to a data driven model also requires commitment from management. Silo organizations are enemies of data-driven approaches. It is therefore necessary to prepare for internal negotiation and adaptation before data collection.
Sources :http://www.creg.ac-versailles.fr/Big-Data-et-Data-marketing#outil_sommaire_1
https://www.fr.capgemini-consulting.com/point-of-view/hyperpersonnalisation-segmentation
