The basis of successful personalization is the interaction of the three core elements: data, content and technology (for delivery and dispatch). Good content is important to get more data. To play out personalized content, you also need the right tools for your own purposes or goals. But without good address and data quality, even the best tool and the most exciting content are useless. And of course, a plan (or strategy) for what you want to achieve with it is also part of it.
But first, let’s take a look at how data contributes to successful personalization.
Become a data collector and user!
Nobody wants to hear the sentence “Data is the new gold” anymore. Nevertheless, anything you know or can learn about your customers can help to provide the customer with a better customer experience through personalized elements. A zip code or geolocation, for example, helps to recommend nearby merchants. Conclusions can also be drawn about the preferences of your customers from the search behavior or the device used.
The customer data landscape and the creation of personalized customer experiences are more relevant today than ever. Many customers expect companies to provide them with content tailored to their interests and wishes.
However, the topic of data protection is now also very important to most customers, so you should communicate your handling of customer data transparently. Satisfied customers who trust a company in handling their personal data also reveal more or high-quality data. This better data quality then creates the basis for appropriate personalization and optimized content.
Silo thinking is a thing of the past
Companies needed to create touchpoints as quickly as consumers increasingly embraced new technologies and channels. In most cases, this led to the introduction of channels and systems that did not integrate the data obtained with them into the existing systems. Data silos have been created with the result that data is confusing, repetitive in different databases and there are no complete data profiles for the respective customers.
To deliver an optimal customer experience, you need to break down the existing data silos. A data audit helps determine which data exists where in your company. Then, you need to set up a strategy that ensures that the data you collect is properly stored, accurate, consistent, and consistent.
To process and integrate newly preserved data into your systems, you need to assess data sources by how well they integrate with your data systems, i.e. whether the data can be imported and exported across systems.
“We expect customer loyalty silos to be one of the top three causes of customer dissatisfaction for businesses across all industry segments by 2020.” – Gartner
Another important point that is often forgotten: data becomes obsolete! So in the sense of an optimal customer approach and personalization, make sure to always keep your data as up-to-date as possible. Because every year about 10% of the data becomes obsolete!
We know this from a reliable source
Data can be obtained in different ways: the customer can give it to you voluntarily, the company can collect or record it itself in dealing with the customer or buy it in. Data ownership – as in the first two cases – is advantageous in that you can process the data as you like. With regard to the GDPR, the possession of the data is also an important prerequisite for the deletion of data or, for example, to ensure “the right to be forgotten”.
Currently, four different types of data are distinguished in the context of data collection: zero-, first-, second- and third-party data. All are based on different acquisition methods and vary in scope and accuracy.
Third-party data – the dwindling source
Third-party data is purchased, often aggregated data, i.e. a large amount of data collected by third parties is combined into a single value. As a rule, this data is not personal. Third-party data can vary greatly in your data quality. So far, however, they have been of some importance when companies wanted to obtain information about non-customers. Market analyses, for example, can be carried out on this database. They also provide all the information that a customer, prospective customer or business partner does not want to disclose about themselves, such as risk indicators on finance or compliance.
The European Court of Justice, the GDPR and the new Telecommunications Telemedia Data Protection Act (TTDSG), which came into force in December 2021, have already strongly regulated the use of third-party cookies. While these could previously be set without the user’s consent, users must now explicitly agree to third-party cookies or cookies for advertising purposes through the so-called opt-in function. After Google declared in 2020 that it would discontinue support for third-party cookies by mid-2023, and the Firefox and Safari browsers have already integrated anti-tracking methods, a large part of third-party data will be eliminated in this context in the future.
Cooperation provides second party data
Second party data refers to data that is obtained from another provider. It can be either a one-time purchase or a long-term partnership in which data is exchanged.
Ultimately, it is first-party data from third parties, such as business or marketing partners, and the amount or quality of this data can vary greatly depending on its origin. For example, data is collected by media agencies as part of campaign deliveries, which can then be provided by the media agency as second party data. This second party data allows you to significantly complete the profiles of users. Above all, this is information about areas of interest as well as other, personal data that you may not request directly from your visitors because you are prohibited from doing so by data protection law.
The advantage of buying second-party data is that you can tell the seller exactly what data is wanted and what is not. A separate, possibly time-consuming collection of this data is no longer necessary.
Something of its own! ꟷ First-party data
First-party data is information that you passively collect from user interactions. First-party data comes from your systems and platforms, such as your website, app, or social media pages. Typically, the most efficient way to collect first-party data is through a Customer Data Platform (CDP) or Customer Relationship Management System (CRM).
The difference between first-party data and zero-party data comes from how users give their consent. While first-party data is passively collected through customer interactions, users intentionally share zero-party data.
One to zero for you ꟷ Zero party data and its difference to first-party data
The term Zero Party Data was first used in 2020 by the American market research company Forrester and defined by them as follows:
“It’s data that a customer consciously and proactively shares with a brand. This can include preference center data, purchase intent data, personal context, and how the person wants to be recognized by the brand.”
In contrast to first-party data, which is collected passively, i.e. usually not visible to the customer, zero party data is proactive. Consumers choose to give you, as a business, a certain type of information, such as their name, email address, and phone number. In the exchange of these values, they expect an equivalent value. This can be, for example, a time-limited offer or a test or demo version of a product or service.
Zero-party data is thus information that a customer shares directly with you about himself. Ergo: These are the best or “hardest” data. One of the reasons he does this is because he wants you as a company to get in touch with him. You can assume that the data quality is good and the information is correct. After all, which customer indicates his preferences in the Preference Center in such a way that he subsequently receives content that does not interest him at all? Since he provides this data voluntarily, there are no problems with regard to data protection. The customer thus has full confidence in your company!
Data strategy as part of the personalization strategy
As part of the personalization strategy, you should sift through the data already available in the house, collect the existing data points from the – presumably – numerous systems or platforms and, at best, already organize them. Keywords are SDP and CDP.
Your future data collection and management in your company should be unified, standardized and coordinated. As already mentioned: Do not build data silos or dissolve existing silos. The data points must be able to be brought together to get a holistic picture of your customers.
Very important: Now consider what you can use of this data for personalization and which you may still need and need to collect for an optimal customer experience. Create routines of how you want to use the data. The following also applies here: In the beginning, less is more.
Data and personalization – this is the reality
In fact, most businesses still use a combination of zero party, first party, second party, and third party data to get information about your customers and audiences. In the future, the first two will become increasingly important in terms of data quality and data protection for the reasons already mentioned.
According to a recent study by absolit, 2/3 of those surveyed use first-party data for personalization, almost 45% use zero-party data and a good third combine both types of data.
However, preference centers, through which customers can store their personal data and preferences, are hardly used yet, as another study by absolit shows: Just 13 percent of the companies surveyed offer one and ask for interests about it – but without the possibility of being able to change the information afterwards (source: absolit E-Mail Marketing Benchmarks 2022).
The fact that it makes sense to think specifically about which data you need and want to or can evaluate was also shown in the study by absolit. Of the 13 data points queried there, the companies surveyed used an average of only 4.5 for personalization.
Comparison of the four types of data
Note: This is a machine translation. It is neither 100% complete nor 100% correct. We can therefore not guarantee the result.