Retaining the right customer with AI: our pitch for the first maritime accelerator

Confronted with a saturated market, shipping companies have to pull out all the stops to expand their slice of the transportation pie. Add increasingly smaller margins and you know they have it rough. For our entry in the PortXL Antwerp accelerator programme, the first maritime accelerator, we wanted to provide an answer to this challenge. Our battlefield of choice? Customer retention.

The question that begged an answer

With many players in the same space, competition is cutthroat. In the blink of an eye, a client can take his business elsewhere and your profits take a hit. To get or keep that big slice of the pie, retaining customers is key for shipping companies. Yet we wanted to take it one step further. It’s not just about customer retention. It’s about keeping the right customer. The one who brings most business and is easiest to work with.

Our pitch

For the PortXL programme, we pitched our approach to getting a grip on the customer journey and retaining the right customer. Our ambition was to shoot for the stars: help port companies increase net profit with 20 percent by reducing customer churn by five percent. We looked at artificial intelligence (AI) to help us provide real-time, personalized customer insights. These understandings would allow us to more accurately manage the experience by interacting with a customer’s journey at the right time according to your client strategy.

Our strategy paid off. The Mediterranean Shipping Company (MSC) was impressed with our plan to conquer customer churn. In the coming months, we’ll be hard at work to bring our idea to life for them!

Customer analytics as the answer to the question

For this use case we focus on creating a customer analytics model with AI. Like a true customer buddy, we made an algorithm that follows client’s along their journey. Based on the interactions with customers logged in the port companies’ systems, it then detects when customers are on the verge of jumping ship. Account managers are notified when this happens, allowing for intervention. Based on company strategy or actions in the past, the AI can even suggest possible courses of action.

Of course, you have to determine which customers are the ‘right’ one to keep. AI can help, for instance, by cross-referencing data on revenue share of a client against number of interactions or container bookings. Take customer X who brings in two percent of revenue and requires a full-time equivalent to manage. Customer Y who has a 3 percent share, needs two or even three people to handle. Even though X brings in less money, he’s much easier to do business with than Y. It’s these type of insights that AI combined with machine learning (ML) can easily surface. And it’s just one of scenario of many.

Pattern recognition at the basis of analysis

There’s one sure-fire thing algorithms do better than humans: recognizing patterns in vast amounts of data. It’s this capability that allows them to discover indications that a customer might leave that a human just can’t. A strongly-worded mail is a clear sign that any human can detect. But while a human is on average capable to reason based on maximum four variables, AI is able to take into account hundreds or even thousands. As such artificial intelligence is uniquely suited to discover insights on customer churn and serve them to you when you need them to retain clients.