There are a lot of factors that influence a customer's decision to cancel their service with a company. From emotional factors like mood and satisfaction to more practical factors like the cost of the service, there's a lot of noise to account for.
When a customer cancels, the costs are enormous. The flipside is true as well: Retaining customers is a proven way to boost profitability, as a 5% increase in customer retention can boost profits by 25% to 95%.
Akkio's machine learning algorithms can aggregate all your information and provide a prediction of how likely they are to renew their service based on each factor, as well as what is missing from the equation. For example, they might be satisfied with your service but they're only willing to pay a certain amount for it, thus preventing them from renewing. Or they might be willing to pay more for your service but their mood is low and they're not satisfied with it.
This insight is actionable and lets you identify which customers want to renew and what needs to change in order for them to do so. You can also use this information to upsell them on a new product or service if they're close to canceling.
To get started, sign up to Akkio for free. As with any machine learning task, the first step is getting historical data and picking a column to predict.
We’ll use a historical customer renewal and churn dataset from a tours and travels company, available on Kaggle. This dataset includes just under a thousand rows, and features just 7 columns:
Each of these columns is potentially indicative of customer behavior, and each will be given a corresponding weight in the model. For example, perhaps a high-income customer is more likely to renew, even if they haven't synced their account on social media.
On the other hand, a low-income customer that has frequent flyer status may be more likely to renew regardless of whether they've synced their account on social media. These sorts of relationships in the data are automatically discovered by the model.
Crucially, the file includes our desired target variable, or whether the customer renewed. We can use machine learning to predict customer renewal, and then prioritize efforts accordingly.
There are many factors that could affect renewal. With Akkio, you can build custom models to address the specific drivers of your business. And with our drag-and-drop interface, you can build any predictive model you want. First, we’ll just download the dataset as a CSV, and upload it to Akkio in a new model flow. You'll get a preview of the dataset, including modifiable data types, columns, and row count.
Now, we can click on the second step in the AI flow, which is “Predict.” Under “predict fields” you can select the column to predict, named “target.”
You can add or remove as many columns as you’d like, and we'll build a model based on the selected columns. Then just hit “Create Predictive Model,” and you’re done.
You can select a longer or shorter training time—ranging from 10 seconds to 5 minutes—for potentially more accurate models. Keep in mind that longer training times will not always necessarily perform better.
Also note that you don’t pay for model training time, unlike with many typical automated machine learning tools, so feel free to build as many models as you’d like.
After a model is created, you get a simple overview, which highlights the top fields, model prediction quality, sample predictions, and more.
For instance, we can see that our total accuracy is around 85%, with solid predictive accuracy for both classes: Customers that renew, and customers that churn.
There’s also the option to “See Model Report,” which lets us easily collaborate with others and share these model details with anyone, even if they don’t have an Akkio account. In short, we’ve discovered that it’s very easy to build an accurate model to optimize inventory.
Now that we’ve built a model, it’s time to deploy it in the real world.
With Akkio, it becomes trivial to deploy complex machine learning models. Deployment via web is an easy way to instantly serve predictions. It’s also possible to deploy through Salesforce, Google Sheets, Snowflake, and the Akkio API, with many more methods coming soon.
Further, you could deploy the model in practically any setting with Zapier, a no-code automation tool that connects tools with “Zaps.” Zapier allows you to link up thousands of different tools, so no matter where your data is coming from, it’s easy to connect it to Akkio to get a prediction, even if you don’t have any technical expertise.
If you’re looking for more technical power, you can also use Akkio’s API, which is formatted as a Curl command, allowing you to send a GET request that contains your flow key, API key, and input data, and get a prediction back.
We’ve explored how renewal and churn are impacted by many different factors, such as the customer’s satisfaction and their willingness to pay. We can use this data to make predictions about each customer based on their characteristics. Additionally, with Akkio, we can create models that update on new data, as customers change their behavior.
By using Akkio, we can easily create a report that describes what each customer’s likelihood of renewal is. We can also prioritize our efforts based on the customers most likely to churn first. This saves you money in the long term, since it's costly to acquire new customers.
There are many other machine learning use-cases outside customer renewal, such as sales funnel optimization, customer support prioritization, or even fraud detection. Using Akkio to gain insights into your data can help you gain a competitive advantage and make better-informed decisions.