To have a well thought-out digital strategy, marketing must rely on data. Discover in this article how data is present in marketing. Know your customers better.

What is data marketing? 

To understand consumers' intentions, marketing will rely on data: this is data marketing! To give a more concrete definition, here are some elements: Data marketing or data-driven marketing refers to marketing based on the use of data. The data will be collected en masse on the Internet. The goal is simple: to transform prospects - i.e. potential customers - into customers. 

We can also talk about digital marketing. 

What will data marketing be used for? 

Here are several issues that data marketing can help address: 

• Measuring ROI (return-on-investment): You put 1 euro into a marketing campaign, how much does this invested euro bring you back? Data helps to answer this question. To do this, you need to know how to measure the impact of a campaign and that's where data skills are needed. 
• Experimenting with marketing campaigns: With the help of A/B testing, we can understand which are the most effective ways to reach our audience. To put it quickly, A/B testing is a way to test your advertising campaign in real time: you present an advertisement A to a group of people and an advertisement B to another group. You then look at which ad is more effective and learn from this to improve continuously. 
• Transforming data into customer knowledge. We will make analyses to better know our target consumer. With the help of socio-demographic data, we will then be able to target more precisely the audience most interested in our product.
  

What are the uses of data marketing? 

Data marketing has many applications that we will detail next. Generally speaking, it will be used to predict consumer behavior as well as possible.

Create targeted ads

As we showed with Netflix and the posters that it can propose in a different way to users, it is possible in digital marketing to propose personalized ads according to the users' path. 

For example, we can do some simple first things. If we know that the user has already bought something, we can offer him, via ads, products that are related to what he has bought. The person has bought a phone, it is then relevant to propose him for example a case or an external battery. 

If we know that a person has previously clicked on an advertisement but has not bought, we can then offer him the same product but with a discount to encourage him to buy because we know that he is potentially interested in the product. And in this case, we assume that the price is one of the factors that made them not buy.

If the person has never clicked, it is potentially that he does not know the brand, we will then propose a generic ad to discover the brand. Potentially, it is also simply that they are not interested. 

Optimize ads

Let's stay in the previous case where the person has never clicked. So how can we optimize the probability that they will click? We can optimize 2 factors: the time of the ad and the content. 

Concerning time, 2 points are important.

• How long should the ad last? For example, when it appears at the beginning of a video, is it necessary to make it last 2 minutes or is 30 seconds or even 10 seconds sufficient? 
• How many times should the ad be displayed? Very generally, if a person clicks, they won't do so the first time they see the ad. On the other hand, showing the ad too many times will be counterproductive for the brand and may make the user distrust the ad. Capping is usually set up, i.e. a maximum number of ads displayed to the same person. 

And of course, you can optimize the content. A simple example: you can play with the color of the advertisement and find the one that best suits your objectives. 

Improve customer relations 

To improve customer satisfaction, we can optimize elements such as: 

• The content of the message
• The time of the message
• By the right communication channel (SMS, Mail, Chat, ...)

In a world where users are constantly receiving information, it is important to send the right message at the right time that will grab the consumer's attention. The content of the message is important, it must seem personalized. 

Whether on the internet or in a physical store, using behavioral data, one can for example propose personalized promotions to maximize the probability that the customer will buy and be happy to buy. 

The first level of personalization is customer segmentation, which consists of dividing all of its customers into a number of defined groups and proposing to each group of customers the offer that best corresponds to this group. 

‍

Improve sales performance

Finding out who among the prospects are most likely to buy is also part of data marketing. We will be able to evaluate the probability that each individual will buy. From there, we can make marketing investment decisions. In particular, if we decide that the probability that the person will buy is high, we will implement the best actions to maximize this probability. 

Once they have purchased, we have a lot of data such as what they buy and how often, how they buy it and where. We will also have the possible history of incidents with the company. 

The after-sales service will therefore be able to provide relevant answers to users' questions thanks to the knowledge of what they have done in the past.

Risk of attrition

On the other hand, data will also seek to reduce the risk of attrition, i.e. losing customers. We will therefore use data to 

• Understand the negative points raised most often by customers 
• Monitor consumer reviews and see if there is deterioration or improvement
• Monitor competitors' offers. 
• Look at the average number of purchases over a given frequency

While it is perfectly sustainable to replace lost customers with new ones, it comes at a cost, usually higher than keeping your current customers. Therefore, it is necessary to try to keep your customers. 

Cross-selling, up-selling and data

We also have, before the sale, the cross-selling: selling other objects in addition to those that the person has chosen, to increase the turnover and the margin. This is typically the coffee or the digestif that the restaurant owner offers you at the end of your meal, but with data to do it to thousands or even millions of people. Downstream, we also talk about up-selling. Amazon, for example, will propose it in the form of "products frequently purchased together". 

Once the customer is satisfied, he can be offered additional features or a higher-end product. This is called upselling. This is the case of the garage owner who sold you a car and then offers you additional maintenance. Generally speaking, this is a very common practice in all subscription-based systems and is often presented in the form: "You like our product? Then unlock this feature that will make it even more useful, for a 20% higher price". Another example of up-selling: extra GB for your mobile subscription. And if the data marketing is well done, this offer will be proposed to you when you will have exhausted your whole package or you will be close to exhausting it.

Optimize logistics

Sales forecast

Thanks to the data, we can make sales predictions. This will allow to evaluate the potential turnover and margin but it also has an impact on the inventory management. From a marketing point of view, the next step will be to add to the predictions the impact of marketing. If we put x euros in one advertising channel, y euros in a second and z euros in a third, what will be the ROI? How much will it bring in?

We will try to make predictions at different levels: short, day to day and medium / long term. 

Inventory management

When selling physical products, inventory management is essential. Correctly estimating the day-to-day sales of each product allows you to save costs. Indeed, buying products that will not be sold obviously represents a cost. But also, not planning enough products will represent a cost in the long run: delivery times will be longer or some orders will have to be cancelled, decreasing in both cases the customer satisfaction. Moreover, planning too much stock will have a cost: stocking is expensive, especially if the storage place is not big enough and it is necessary to expand.

Conclusion 

Optimizing ads and creating targeted ads, improving sales performance by generating new customers and knowing how to keep them while properly forecasting inventory and keeping the customer happy are all use cases of data in marketing which makes it a very important field of application of data!

‍