AI for B2B Marketers – Practical Use Cases

Database hygiene the bugbear of every marketer. The unbearable stench of rotten leads or contacts – those that are patently fake, missing required fields, or of ex-staff.

Cleansing these contacts is the stuff of nightmares. But AI/ML can give some respite. We ran a 6,000+ database of email contacts acquired from past campaigns, events and such through our Python-based LeadCleanse algorithm. About 35% came clean, saving hours of manual verification. What’s more, LeadCleanse augmented valid contacts with essential data such as Job Title and Company Name that were missing from the original data set. With LeadCleanse, we can validate email addresses for deliverability, and phone numbers using telco APIs for reachability.

What’s the correlation between LinkedIn spending and revenue per client? Will increased media spending on trade shows lift sales volume, or affect customer lifetime value? What’s the impact of Net Promoter Score on the bottom line?

Media budget allocation is mostly driven by ‘gut feel’. Machine learning regression analysis (MLRA) can help a marketer predict the mix of outreach channels that would lead to a higher revenue pipeline. MLRA is a set of machine learning methods that allow us to predict a continuous outcome variable based on the value of one or multiple predictor variables.

For telcos, ISPs and SaaS providers, identifying customers who are more likely to “churn”, that is, quit the service after contracted periods is something of a black art. Often, they heap considerable effort and expense upon those who least likely to churn, while those who do churn inexorably slip past the cordon. Furthermore, and perhaps to the detriment of customer lifetime value, loyal customers fail to be rewarded for their loyalty. ChurnSense is our Customer Churn prediction model that predicts which customers are about to churn, and target retention campaigns at the customers most likely to churn – with up to 78% accuracy. The same model can help identify loyal customers who can be rewarded to cement their loyalty for longer periods of time. ANN, by the way, is not our wizkid in the basement who works on ChurnSense (though she could be). Artificial Neural Networks (ANN) is that component of AI that is meant to simulate the functioning of a human brain. Processing units make up ANN, which in turn consist of inputs and outputs. ANN learns from these inputs to predict the output. And customer churn production modelling (what we call ChurnSense) is just one of the many use cases of ANN.