Lead nurturing, lead scoring, dynamic segmentation and triggered communications are much more than pure analytics worth taking a look at to identify the success factors and potential pitfalls of marketing automation:

1. Consolidate for an all-around inspection of the customers

As a basis for marketing automation, companies should merge internal and external customer information and transaction data from different sources into a data warehouse or data mart (CRM, ERP, web shop, email service provider, excel spreadsheets, purchasing power data, etc.). Start with the two or three most important data pots and integrate other sources gradually thereafter. Provide clean and current data in order to be able to select granularly - for example, master file data is assigned to each transaction data set and vice versa, and duplicates are matched and adjusted by algorithms. In addition, marketers have to get a general idea of data records with permission-based email addresses.

2. Get granular on business data and customer behavior

Call it Big Data or not, fact is that with just 10,000 customer master data sets, the number of transaction records can quickly skyrocket into the millions. Obviously, the music plays precisely here. Put your attention on the transaction data and derive from the customer behavior rules – such as e.g. the optimal email frequency of the newsletter for each customer. Selections, visualizations and statistical tools for scoring and customer profiling help of course, too. However, please keep in mind to start with marketing automation only after you have created the analytical basis for it. Only good preparations guarantee effective targeting for successful campaign management - especially when it is controlled automatically.

3. Manage type and frequency of marketing impulses properly

Conduct a revision to determine which of the many possible campaign operations generated a positive contribution: Once set up, dozens of campaigns can easily run fully automated at the same time. However, make sure to limit your communications temporally and across channels in order not to deter customers with numerous impulses. Instead, determine a capping, such as for example "no more than one email per week", and limit the number of mailings to one person per household. Besides that, consider event triggers - such as rule-based "customer has clicked"; time-based "client did not respond within 48 hours" or response-based "customer bought". Only then, testing of the control groups makes marketing automation practicable.

4. Lead Nurturing

Actually, only a fraction of the potential buyers is prepared to purchase; the rest is in the research phase, usually feeling bothered by too frequent communication and offers, so you risk that they back out completely if you are too pushy. Hence, lead nurturing comes into play, helping to keep the parties interested. The focus is on customer loyalty, sending relevant information to the customer, at appropriate frequencies via the preferred channel. The right contact strategy, along the customer journey to the point of purchase, can be derived from open- and click rates among many other options. Simply keep in mind not to overwhelm those that read just infrequently due to loads of emails, as it annoys them and you risk unsubscribes. In reverse, thanks to optimal timing and channels, responses of automated campaigns are attractive and lucrative.

5. Learn from past campaigns

Via interfaces, for instance to email service providers, personalized emails can be sent automatically and their opening-, click-, and bounce rates can be integrated and analyzed almost in real-time. Feedback integration applies to affiliated CRM systems, call centers, and mailing responses which have just been captured. The associated transaction data is mainly flowing back into the process through a source system such as a CRM database. And last but not least, try to promptly trace back the responses and evaluate the results.

By Daniela La Marca