Part 1: Using data aggregation in your DWH to improve your marketing strategy
In the first part of this two-part blog post, our Director Perfomance Marketing Dorothee Seedorf shares insights on customer segmentation and how using data aggregation in a Data Warehouse can improve your marketing strategy.
Every recently launched startup with a transactional business model needs to overcome a certain threshold regarding its customer base in order to be profitable. This means acquiring new customers and converting users as well as retaining repeat customers. The latter in particular is becoming more and more important in light of today’s advertising landscape. Cleverly orchestrated customer relationship management (CRM) decreases marketing costs, supports the sustainability of your business and even enables differentiation and thus a competitive advantage. We support our ventures in integrating a professional CRM process into their marketing strategy right from the start and help them to establish suitable customer segmentation in advance. An important factor is to create personas and specify target groups in order to define your business goals beforehand. This lays the groundwork for the subsequent customer segmentation, which is essential for any properly executed CRM.
The quality and sophistication of CRM measures mainly depends on two factors: first, the available data at your disposal and, second, the underlying models which determine how to utilize this data.
To avoid shortcomings and to prevent redundant aggregation of data, it is important to assess beforehand, or at least at a very early stage in the process, which data you will need later on. There is no bigger pain than having your data warehouse (DWH) crammed with customer data you cannot use at later stages. The same goes for data you do not save but end up needing in the future.
Before any data gets aggregated, your budget and setup of marketing channels need to be defined in accordance with relevant target groups. This process is managed fairly by KPIs like Cost per Acquisition (CPA), Cost per Lead (CPL), or Customer Revenue Relation (CRR).
The main purpose of aggregating customer data is to gain actionable insights that lead to a higher conversion or repeat purchase rate. The more granular the marketing measures, the better you can react to specific customer profiles. But before you are able to personalize your ads, you need to segment your customer base. The level of granularity needs to be assessed with respect to the composition of your customer base, your product offering and the costs. The segmentation of customers requires a counterpart in your marketing strategy that in turns depends on the products you have to offer. There is little sense in having a fine-grained customer segmentation without having the means to personalize your ads for each segment. This would unnecessarily increase the costs without providing you with any additional advantages. It is all about finding the right balance.
What kind of data and corresponding segments are we actually talking about? In general, any data helping you to differentiate your customer base could be relevant. But which data turns out to be useful depends on your specific business model and customer needs. Data that could be aggregated includes:
- Demographics (age, nationality, religion, gender, income, educational level, etc.)
- Geography (residence, job location, etc.)
- Psychographics (social class, lifestyle, personality characteristics)
- Behavioral data (data from customer journey: interest, buying intent, purchases, repeat purchase rate)
Analyzing your customers’ interaction throughout the customer journey enables you to learn something about your customers and thereby improve your marketing opportunities. Dividing customers into different groups of specific interest, age, gender or spending habit will provide you with a more effective way to utilize your marketing resources. And this does not only apply to B2C models. B2B businesses can also profit from customer segmentation and personalized ads, even though CRM processes work a little differently in this case. Often they are more complex, since decisions are not made by individuals but by several decision makers instead.
Our venture Contorion, an online store for industrial and trade supply with a B2B business model, is an interesting example here. Classical marketing measures are only part of the strategy in this case, because customers are older and therefore have less affinity with the internet. This is why they in general prefer traditional marketing channels such as phone, fax or catalogues. Since they are very loyal customers with a high customer lifetime value (CLV), even more expensive CRM measures are economically profitable.
No matter which business model is involved, there are always various possibilities regarding customer segmentation. It is not always obvious which segmentation will perform best. Establishing user cubes for cohort analyses helps to identify how each group is performing. Depending on the defined cube, the way you interact with users might change, e.g. regarding dimensions like channel, content, time, frequency and budget worth spending.
So much for now! In part two of this series I will provide you with selected examples on how to utilize user data in your marketing efforts.