7 Steps Subscription Brands Can Take to Reduce Churn
The subscription e-commerce market continues to see rapid growth, with McKinsey’s latest figures suggesting it now totals between US$12 billion and US$15 billion in the US alone.
Alongside the big global players, such as Hello Fresh, the Dollar Shave Club, and Naked Wines, are an increasing number of smaller, more localized brands such as LookFantastic, Pint Society and Japanese furniture and interior goods provider Muji.
Despite the growing popularity of brands such as these, subscription companies typically face high churn as they struggle to keep customers over time. So, what can subscription brands do to prevent and reduce churn?
What Is Churn?
Churn is the percentage of customers who stop using your products or services within a specific time frame. According to Recurly, a global subscription billing management company, the overall churn rate for subscription brands is 5.6 percent with subscription box and consumer goods running slightly higher.
Customers churn for several reasons. A big one is to reduce expenses – an issue that has become more prominent since COVID-19 as people have become more conservative with their spending. Other reasons include less frequent use of a product or service, regular travel, and the seasonal nature of some subscriptions.
According to McKinsey, 40 percent of people who subscribe to a product or service go on to cancel it. In addition, more than a third cancel in three months or fewer. As churn can greatly affect profitability, minimizing it is key to long-term success and growth.
How to Minimize Subscription Churn
To reduce churn, subscription brands need to continually connect and demonstrate value to keep customers paying up each month, quarter, or year – something achievable with the right tactics and the use of massive data and AI-driven technologies.
1. Predict potential churners to rebuild a connection
Any customer who has signed up for subscriptions can potentially cancel at any time, but some customers are more likely to cancel than others. If you can figure out who these customers are, you can work harder to make them stay.
Leverage predictive machine learning (ML) models to analyze usage and contextual data from your existing customers. Usage data shows how much a customer used your product or service before they left, while contextual data adds more context to the former. ML can strategically predict which customers are high-risk based on their actions. For example, people who recently downgraded their subscription or who have been inactive. You can then engage them with regular, relevant communications to reignite that connection.
2. Ask customers why they want to cancel
Understanding why people want to cancel a subscription is crucial. It helps you create a strategy to avoid similar future losses. For example, if people are leaving because your service is too expensive or they aren’t using your products, you could look to adjust your prices or rethink your offering.
You can collect this data via a simple online survey, by email, or over the phone at the point of cancellation. Make it easy by giving a few multiple-choice options, plus give them the opportunity for open feedback. At the same time, tackle those pain points by suggesting they shift to a lower tier or pause their subscription temporarily.
3. Develop real-time customer profiles to stay relevant
One reason customers cancel subscriptions is that they no longer have the same level of need or interest in that particular product or service. This can be avoided by adopting a data-driven approach.
By leveraging legally-compliant third-party data on customer behavior and interests outside your owned channels, you can use AI tools to develop more accurate customer profiles in real time. This will provide you with a better understanding of your customers’ changing needs and preferences, so you can take actions to ensure your products or services remain useful and relevant.
4. Bring back lost customers based on lifetime value
Even if customers cancel their subscription, it is never too late to bring them back. By targeting the customers most likely to re-subscribe with engaging communications and relevant offers, you can see better success and increase your ROI.
To identify these high-value segments, you can use deep learning tools to define specific segments based on the audience’s massive behaviors and interests. You can then use predictive deep learning models to combine the best segments based on predefined rules and rank the segment combinations based on their value or likelihood of re-subscribe.
5. Re-engage through the right channel at the right time
Preventing churn is all about keeping your subscribers sticky – and to ensure that, you want to stay visible, connected, and continually deliver great experiences.
Using AI-based marketing automation, you can better understand your subscribers’ preferred channels of communication (email, app push, in-app messages, SMS, messengers, etc.), their top times of day for engagement, and what types of creative formats they respond best to. You can then send personalized, interest-based offers to re-engage.
6. Explore partnerships to provide added value
Partnerships are the new black in retail, and subscription retail is no exception. Teaming up with another brand can not only help your bottom line, but it also allows you to offer subscribers something new and desirable – increasing re-engagement and reducing churn.
For example, an ice cream subscription box (e.g., Pint Society) could partner with a movie streaming service, or a clothing subscription brand (e.g., Her Velvet Vase) could team up with a beauty brand. Instead of simply guessing which partnerships might be valuable for your subscription brand, leverage AI-powered customer insights to explore ones that could provide added value to your subscribers.
7. Reward subscriber customers for their loyalty
The more loyal the customer, the more they value what you offer, and the less likely they will leave you. Because of this, incentivizing and rewarding your best customers, as well as your new customers, is crucial.
To boost loyalty to its brand, food subscription box retailer Hello Fresh offers customers discounts and a free box for referrals. Birchbox does the same by offering members 50 points for each referral. You could also incentivize other activities, such as engaging on social or watching videos.
While a certain amount of subscription churn is always going to happen, you can keep it down and profits up by using data and the right tools to better understand what your subscribers want and why they come and go, and work hard to stay connected and deliver value.
* Learn more about how machine learning and marketing automation can help you reduce subscription churn. Contact our team today to arrange an exclusive consultation.
WE ARE HERE TO HELP
YOU MIGHT ALSO LIKE
For any marketer, one of the first and most critical steps in any successful new product launch is identifying the right audience. Rather than relying on demographics and guesswork, leveraging artificial intelligence (AI) can help you find the prospects with the highest potential to convert among your existing customers, and beyond. During a new product launch the pressure is on. Not only do you need to come up with a sound marketing plan, you also need to work within a budget and demonstrate good ROI. Before you get into the detail and tactics, the first and most important step in any product launch campaign is identifying the right audience – the potential customers most likely to engage with and buy your new product. This is typically done by figuring out the ideal audience profile for your product. For example, if you are launching a new top range razor, you might narrow your audience down to males, aged 25-45, who like personal grooming and have a high disposable income. However, by talking this generalized approach it is hard to guarantee, with any level of certainty, who in this group will engage with or buy your product. In addition, if you only
Every consumer is different. They have their own interests, preferences and concerns. Sending the same message to every one of your customers and prospects is unlikely to win their hearts. Instead, it will only see your efforts quickly ignored and leave a sour taste in their minds. For marketing to be effective in any industry, you need to find a way to speak to your audience on a personal level, and using personalized techniques backed by artificial intelligence (AI) is the way to go. Marketers’ Missing Opportunities for Personalization It is now a common practice to use marketing automation tools to reach a wider audience, but the most obvious mistake that marketers tend to make is simply ignoring the option of personalization. By monitoring how users interact with your site, you can get very clear signals of what they are looking for, which device they use and at what time of the day. Failing to engage them with personalized messages means you are missing out on a hot lead. Other marketers begin personalization first interacting with the audience, but stop short of tailoring messages to the individual throughout the customer journey across devices. For instance, a visitor looked at
Author | Dr. Min Sun, Chief AI Scientist, Appier Artificial intelligence (AI) and machine learning (ML) have moved from the backrooms of computer science into the mainstream. Their impact is being felt in everything from how we shop through to money markets and medical research. Larger models have been trained in separated modality. For instance, GPT-3 is the first 100-billion-parameter model for natural language processing (NLP). Recently, a-trillion-parameter model (T5-XXL) has also been trained. They can be used to write articles, analyze text, perform translations and even create poetry. In parallel, we’ve seen models used for image recognition and generation greatly improved as they have also been trained with more data sets. What we are seeing emerge is the power that can come from combining two or more AI models without changing these large models. In this way, combining these large models becomes affordable. That will allow us to use AI to interpret text and generate a completely new image. We are also seeing how the architecture of one model can be adapted to solve problems across different domains. The most powerful example is how the architecture that powers NLP models is being used in biomedical research. In the biomedical