FROM PROF. DR. PETER GENTSCH

How you can sustainably increase customer lifetime value

Huge sales potentials are given away. Because: Sales has changed and continues to change rapidly. Your customers – whether B2B or B2C – expect and need a much more targeted approach, at the right time and on the right topic. Without this, sales today quickly become a turning point for the entire company and a clear competitive disadvantage. Studies show that classic CRM is losing its importance in this context and that much more dynamic approaches are coming to the fore: real-time data from analysis systems, integrative analysis of various channels and, in particular, predictive approaches are the ways forward for sales in this decade.

If existing customers are taken over by a new sales representative, this is associated with a high risk. Sales often still functions in a very personal, little standardized way and success is purely the responsibility of the respective employees. A new sales manager must first familiarize himself with his new customers before he can process them on the same level as his predecessor.

But why actually? Regardless of size, every company generates more and more data points on each customer and thus has the information to better understand existing customers and to win new customers more effectively. Or in short: to improve new customer acquisition and significantly increase customer lifetime value.

What has long been part of the daily routine in online marketing is also possible in sales: the individual approach to prospective customers and customers depending on their behavior based on the behavior of the entire customer base. So if you know who your customers are and how they behave, you can evaluate these trends and then derive concrete actions to increase sales performance – fully automatically or as a trigger for the sales process. The sales process can thus be standardized, allowing all customer contacts to be processed with consistent quality – regardless of the employee.

The first step is often to identify real commonalities in the data. After all, traditional clusters, such as industry sectors or sales figures, ultimately say nothing about the real similarities of prospects, customers and contacts. The available data is already very valuable here: as in online B2C retailing, precisely tailored recommendations can be generated quickly – not only with regard to the products and services to be recommended, but also at the exact time of contact.

In general, data-based working methods require a departure from traditional opinions and categories. All previous classifications are only crutches, in order to have certain options for action without data analysis. And such “man-made” criteria all have one thing in common: they are very distorted, not very objective and greatly simplified.

The classic CRM with its manually recorded data is at most still a basis or a tool, but no longer the driver for sales. This data-based approach has long since ceased to be the preserve of large companies and corporations, and now offers comprehensive opportunities, especially for SMEs – in both the end customer and business customer business.

Data-based sales processes offer many advantages, starting with a significant increase in the transparency of the entire process, through better planning of sales pipelines through predictability and standardisation, to increased customer satisfaction.

THE SIX MOST FREQUENT ERRORS IN DATA-BASED SALES PROCESSES

Especially at the start of the use of data-based processes, mistakes are often made, which often nip success in the bud. Because if the wrong data is used, if it is evaluated incorrectly or if the use of the results is not thought through, data-driven sales cannot be successful.

Not trusting the data

If you want to work data-based, you must also work data-based. Often the results from the data are counterintuitive, i.e. they do not correspond to what one would have expected from experience. Exactly here lies the biggest chance, but also the biggest risk of the data-based approach. The results must be trusted. Adjustments and fine-tuning must be made to the database and the algorithms. Under no circumstances should the results be used selectively. This leads to at least two problems: first, the results will not be as expected, which weakens trust in the data. Secondly, unstructured processing of the results leads to incorrect feedback to the database, which makes it impossible to optimize the evaluation methods.

Generalize data

Do you still use criteria such as “industry”? Industry is an outdated generalization that makes it extremely difficult to classify companies in a meaningful way. Hardly any company can be clearly assigned to a classic industry – and even if it can, it is usually only superficially correct. In other words: the more detailed data is collected, the better it can be evaluated. The classic generalization to “industry”, for example, shows how such generalizations weaken data quality and thus reduce the usability of the data. Similarities in the data are easy to find – there is no need to summarize or generalize data in advance. “Industry” should therefore be abolished like many other generalizations and replaced by much more granular aspects.

Do not include all data

Data-based work works particularly well when all existing data is linked together. Thus, call data from the telephone system, payment history from accounting, availability of goods from merchandise management and similar data sources are good additions to the classic data in CRM. It is also important to use as little manually entered data as possible. The more automatically the data is generated, the higher the quality of the data. Because even if errors are made in the automatic generation of data, they are structured and comprehensible – and thus the errors are either correctable or not negative for the result of the evaluation.

Do not observe data protection

Not least due to the implementation of the DSGVO, companies in Europe have become more aware of data protection issues. Accordingly, the partner is expected to take this data protection seriously. If automatic e-mails are now suddenly sent without consent or data from different systems are compared with each other, this can not only lead to irritation, but also to lengthy discussions and problems with existing customers. Precisely because many tools come from the USA and there is a completely different understanding of data protection there, it is important to pay close attention to the implementation of data protection requirements when introducing them. But: one must also make use of the possibilities of data protection. Unnecessary restrictions just to achieve absolute legal certainty are certainly not a recipe for success.

About individual distribution

The self-image of many a sales employee collides with a data-driven way of working. Each contact is addressed individually and according to mood, the selection of the contacts to be addressed is largely based on subjective aspects and ultimately everyone receives a highly individual sales pitch and a completely individualized offer. Even if the long-established salesperson sees this as perhaps his or her greatest strength, this is not an advantageous procedure for the entire sales organisation: only a structured approach and data can generate effective insights.

Data at the heart of customer journey optimization!

In order to be able to optimize the entire communication chain with prospects and customers – online and offline – it is necessary to have as much data as possible about the behavior and interests of the contacts in the entire decision-making process and to link them with each other. In this way, you can find out what the path from the first contact point to the sale actually was – and can be optimized to increase sales quotas and custom lifetime value in the future. By combining data from different systems, synergy effects between touch points can be identified and used to improve the results of the various sales and marketing channels. Overall, this provides valuable customer insights to understand the buying behavior of the target group.

Especially against the background of the growing sensitivity for data protection and current legislation (e.g. DSGVO), tools from Germany are preferable here. Thus, there are significantly fewer legal problems and tracking can be set up more completely without running into legal risks.

In online marketing, for example, one sees solutions that were originally developed as a bid management solution for Google Ads and now combine the possibilities of measuring, understanding and optimizing all advertising channels on a single platform – especially with a focus on tracking, search optimization and the integration of e-commerce functions such as Google Shopping and Amazon Advertising.

In this way, it is possible to identify the customer-journey chains that sell most often and best – and that lead to sales with a particularly high shopping basket value. With these insights, marketing activities can be orchestrated much more efficiently.

The respective “optimal attribution model” depends on the specific business model, the company’s objectives and the complexity of the market and can be individually mapped to meet these requirements. A distinction must be made between static and dynamic attribution models: static models are based on simple rules, while dynamic models are based on calculated relationships. The latter obviously offer much greater potential and require less manual maintenance and development.

These attribution models can be used as a basis for budget analyses as well as for the optimal allocation of the marketing budget, so that the optimal attribution model can be determined individually on the basis of one’s own data and can take into account a wide range of influencing factors when determining the model. Thus, a model with high stability can be found that also works across seasonal variations.

Thus, the optimal use is made of the fact that all data is kept in one platform. Driven by machine learning and self-learning algorithms, the control of the individual advertising channels can be optimized automatically – with individually adjustable targets and weightings of the individual touchpoints in the customer journey. Tracking is not only limited to digital marketing channels and CRM data, but external data – including print and TV advertising – can also be integrated and evaluated in the context of the entire customer journey.

Augmented Intelligence instead of Artificial Intelligence!

What does “data driven” mean in practice for sales and marketing? And what can “date driven” actually achieve? Although the potential benefits of “artificial intelligence” for increasing sales are manifold, there are still very few successful practical examples, especially in Germany. Often the potential for artificial intelligence is reduced to the chatbot phenomenon. However, in practice most chatbots are far from being “intelligent”. Either they use methods that were already implemented in 1966 by Joseph Weizenbaum in his “Eliza” or they even conduct only strictly prescribed dialogues without really reacting to the user’s input.

Companies have to find the optimal balance between automation and human interaction in the concrete application. “Data-driven” ultimately means “AI”. And “AI” is not to be understood here purely as “Artificial Intelligence”, but always as “Augmented Intelligence”. It is therefore not necessarily a matter of completely automating processes and methods and leaving decisions to a computer, but of finding the optimal support and potentiation of human intelligence through intelligent algorithms.

The central basis of AI is the data available or to be generated. This means that customer data is of particular importance in AI-based marketing. The topic of AI and customer data is currently dominated to a massive extent by the so-called GAFA economy (Google, Amazon, Facebook, Apple). The goal of the GAFA economy or BATs world in China (Baidu, Alibaba, Tencent) is to improve the ecosystem of consumers and to be able to operate accordingly. Whoever can best master this task can also best place his own products with the customer. It is not for nothing that the GAFA world develops these systems in order to monopolise access to people.

So it is time to collect this data yourself and not to make it available to the big players. Because even if they act as low-cost service providers with high-quality products in the first step, the data will be used for many other things in the long term. Amazon sells well-running products all at once, Google simply offers frequently requested services, functions and information itself, and other providers also use the data in very different ways.

The own database with specialized providers is a sustainable solution for the long-term success of marketing and also offers the chance for special innovations and competitive advantages over mass providers. After all, this is precisely where the primary differentiating factor lies in the long term: how the data is optimally used to address, guide and support the user in the best possible way.

In areas such as lookalike audits, for example, very good results can already be achieved practically fully automatically – where in other places the data used to create advertising materials and plan marketing strategies only support the people and not replace them. With this holistic approach to AI, however, it is possible to get the most out of the data. No unnecessary energy is invested in automating things that are not (yet) suitable for this purpose and data is still used to make the best possible decisions.

Thus AI becomes the decisive element that decides whether a marketing and sales strategy is really successful and can hold its own against strong competition.

AI as gamechanger for Customer Lifetime Value

Most data-based applications use AI to automate functions and processes. In the context of this automation, systems also make simple decisions on their own. This usually involves replacing human activities with AI to achieve cost and efficiency benefits.

Thus, there are many automation applications that already have a high degree of maturity and use in practice today. These include marketing automation or real-time bidding, for example.

In contrast, augmentation applications are particularly concerned with the intelligent support and enrichment of complex and creative sales tasks that are currently still usually performed by humans.

This enables AI to automatically analyze buying behavior, competitors, target groups, and trends. Sales and marketing managers can use these insights to develop or adapt their strategy. In this way, decision-making processes are enriched with important information, but the actual decision is not automated, but remains the sovereignty of human actors.

In terms of “augmentation”, AI can also help sales to manage the increasing complexity of channels and touchpoints. It can calculate the value contribution of a channel as well as the necessary interaction between channels to optimize the sales result. Based on extensive customer data, the optimal sales and marketing strategy can also be determined over time: When should which Euro be invested in which channel? The final sales concept, including the organizational role allocation, is then – at least still today – created and reviewed by people.

Identify leads based on data

Existing data can therefore not only be used to better serve existing customers and increase sales with customers, but is also valuable for acquiring new customers.

By means of “lookalikes”, similar companies can be found on the basis of the existing customer base – or selected “top customers” – that offer high potential as new customers because they have great similarities to existing customers.

The existing company data can be enriched on the web with over 10,000 data points to form a “digital DNA” and new target groups can be identified using so-called “deep learning algorithms”. The concept of “digital twins” or “lookalikes” is used here, i.e. companies are sought that have a similar “digital DNA”.  Based on given A-customers, new customers and markets are identified via statistical twins. In this way, leads and markets can also be identified that are not in the classic prey schema, but represent potential buyers.

A first step towards pragmatically using the possibilities of “lookalikes” are advertising networks of Facebook and LinkedIn. Existing data is used so that the platforms can find similar contacts based on their own data and display advertisements to them. This works wonderfully not only in the B2C segment, but also for B2B products.

The fact that AI already successfully supports marketers and sales staff in this understanding is shown by the use of this approach by digital innovation leader Ströer, who uses an AI solution for new customer acquisition and marketing campaigns: “We deliberately did not impose any restrictions or targets at the beginning. Just as the AI machine predicted the leads for us, we supplied our sales areas. In some areas, we were as good as the traditional sales approaches right away, in other areas we were even better in terms of wastage and conversion. What is decisive is the training of the AI system through a systematic feedback loop – because even the AI system is not perfect from the beginning, but it can learn from it. Andreas Müller-Leydig (COO Ströer Media Germany GmbH)

By means of intelligent data collection and the use of existing data, the identification of leads can easily be extremely improved. Classical remarketing approaches can be quickly optimized by a data-focused way of thinking so that only high-quality contacts are addressed. This saves advertising budget and makes sales much more efficient.

Maximize customer lifetime value

And ultimately, sales efficiency is one of the biggest levers for maximizing customer lifetime value throughout the entire business relationship. A “watering can” approach to marketing and sales, where all activities are simply distributed evenly across all customers and contacts, is therefore no longer a sensible solution. A data-based approach offers the opportunity to communicate very specifically with those contacts who are likely to be receptive to appropriate marketing and sales approaches at the present time.

THE RESULTING ADVANTAGES ARE MANIFOLD:

• More targeted use of advertising budget

• Significant increase in sales efficiency

• Increase of customer satisfaction

• Reduction of “advertising blindness

• Increase in turnover and profit

Addressing the right target groups with the right advertising message not only ensures greater success of the advertising measures, but also offers the possibility of using advertising budgets in a much more targeted manner, thereby avoiding wastage. The efficiency of advertising measures and the users of these in the respective phases of customer acquisition can be measured and used for optimized advertising planning. The fact that the “advertising pressure” on the individual customer is thus reduced, offers the chance to optimally place the advertising statements that are relevant for this customer, to stand out from the masses and to land neither in the real nor in the “mental” spam folder.

The sales department is supplied with pre-qualified contacts who have a significantly higher probability of closing a deal. Valuable sales capacities are thus used much more efficiently. And due to the higher sales success, an increase in motivation of the sales staff can often be observed.

The consideration of learnings from customer communication within the scope of a “feedback loop” to AI ensures that the quality of the evaluations is constantly increasing. Without feedback, neither an employee nor an AI can learn. The AI can reliably use this feedback to optimize further predictions – also with regard to the desired goal. After all, in addition to the pure goal of “more sales”, there are quickly other goals that can be pursued in a targeted manner using such a data-driven approach: be it the targeted launch of new products precisely to potentially interested customers, the promotion of special offers without generally reducing profits, or expansion into new target groups and markets.

Thus, marketing and sales almost automatically become “customer value based dialogue marketing”, i.e. addressing the customer with the added value that suits him.

Market leader vs. latecomer

In data-focused times, it is not the one who delivers the best product and service that becomes the market leader, but the one who uses his data in such a way that he can communicate optimally with the customer and offer the right solutions at the right time. Sales processes thus suddenly become a sales argument. Because even if this is not necessarily perceived explicitly by the customer, it is a strong psychological factor if the (potential) customer feels that he is much better understood and optimally addressed. And this not only when the responsible sales employee calls more or less randomly at the right time, but based on data, repeatable and scalable. Stragglers are hit twice as hard: not only do sales successes come later, but the company also lacks the insights from the data. This can quickly become a real and threatening competitive disadvantage.

CONCLUSION

Without data-based sales there is no market leadership – and for many companies there is no survival. The data-based approach is clearly superior to the classic, manual procedure in all areas. And much more importantly: automation ensures reliability and plannability. This is an enormous competitive advantage.

The faster the first steps towards real “data-driven dales” are taken, the earlier the results will be visible. These results can be seen throughout the entire company, ensure success at all levels and ensure competitiveness. It is important, however, that “data-driven” is not just lip service or half-heartedly implemented. A data-based approach will not work if it is not thought through or only implemented superficially.

Correctly implemented, however, “data-driven sales” is the secret weapon for dominance in the market – especially for small and medium-sized companies.