Reducing churn is essential to growing any subscription business. Lower rates of churn increase revenue, reduce the cost of acquisition and increase the lifetime value of your customers. Industry research suggests that a 5% reduction in churn can increase profits by upwards of 25%.
The basic playbook for reducing churn is similar across most software and analytics businesses. Phenomenal onboarding, great customer service, and responsive product teams all help give customers a reason to stick around. But data businesses in particular face a set of unique challenges – and require distinct tactics – based on the dynamics of how data products are sold and consumed.
Why churn matters
Data businesses face two big challenges that software companies do not. First, the value of a data product is mostly determined by the quality and uniqueness of the dataset. No amount of work from a product or engineering team can overcome a dataset that is not valuable. If an upstream provider changes the underlying dataset (or your access to it,) there’s not a lot you can do to compensate.
However, it also means that commercial teams need to take advantage of everything else they can control. It makes it even more urgent for commercial teams to make sure they nail the opportunities to build enough loyalty and new opportunities at an account so a change of strategy or or alteration to the underlying data does not end the relationship. In an increasingly competitive environment, data businesses need to make sure that they are doing everything they can to help customers find value in their dataset as fast as possible.
We spoke with sales and customer success leaders at data companies to identify six tactics for reducing churn.
With onboarding, speed is everything. The faster data gets in the hands of end users, the more likely it is that the customer will get value from the data. The challenge is that onboarding is typically anything but fast. According to a recent survey of financial data providers, over half of respondents said it took at least 3 months to go from trial to production.
One of the big culprits: integration and prep. Another study of data buyers found that they spend 70% of their time with a new dataset on integration and prep and only 30% on actual analysis. Overburdened data teams responsible for integrating the data also slow down the onboarding process and reduce the chances of the customer finding that winning use case.
Modern data sharing is an important new capability for data businesses. Initially pioneered by Snowflake, every modern data platform now offers the ability for companies to provide secure, read-only access to a dataset. With a data share, analysts get ready-to-query data instantly in their data platform often with little-to-no support from their data team. Data sharing eliminates the 15 days most data buyers on average spend integrating data into their platform via most API or FTP-based deliveries.
Improve the UX of your product
With data products, your schema and symbology are your user interface. Inconsistent or unintuitive schemas can create friction when end users or data teams start using your data. That friction can slow onboarding substantially and increase the likelihood that your customer will not find that first great use case for your product.
A consistent and intuitive symbology is critical. The development and management of consistent identifiers that enable users to easily link and join your products to other datasets is essential to reducing the time-to-value.
Solve problems early
Stuff happens: errors emerge, systems fail, priorities change. Loyal customers are often made in these moments if your team can react quick enough; and trust crumbles just as quickly if you cannot. The challenge for data businesses is that they historically lacked access to the type of observability (e.g. was the delivery successful?) and telemetry (e.g. how did the user interact with our products?) that SaaS companies can access.
There are two investments that teams can make to get information before it becomes a problem. One is commonly known and the other is just as important, and widely used in SaaS, but historically harder for data providers to obtain.
The first is to make sure you have regular check-ins scheduled with users. Make sure to actively ask for feedback to pull out any potential issues with the implementation or usability.
The second is to get access to delivery and usage data. With Bobsled, providers can see immediately if a share did not reach their customers' environment. They can also evaluate basic usage data through a single control plane to see if a customer has not accessed the data. Observability and telemetry data is a powerful tool that SaaS companies have used for years to proactively reduce churn.
Feedback loop to product
Customers understand that every product has its tradeoffs. But you can build loyalty if customers feel that their opinions are actively shaping how you’re building your product.
Account and customer success teams can help win over at-risk customers by making sure the feedback loop with the product team is transparent, fast and effective. Telemetry and observability data can provide critical quantitative information to support specific product features as you advocate for customers requests.
A system with a single point of failure is never sustainable. If you only have one group using your product, a champion leaving or a group changing strategy could kill your renewal.
That’s why it’s critical to immediately start building multiple relationships within the organization. Ask for introductions to other groups. Host lunch and learns to show what your data can do. Give your champions the materials and prep they need to feel comfortable bringing you into other parts of the organization.
If there’s a central data team, invest in building deep relationships with its leaders to figure out how you can make their lives easier. Investments in improving the usability of your data (e.g. supporting sharing) can go a long way in building good will among these teams.
Always tie it back to value
Sometimes, the value of your data is obvious. But many times, a data product is one input in a complex model. It’s essential that you constantly identify, document and reinforce areas in which your dataset is helping your champion generate value for their business (and eventually get promoted.) That means not only identifying when you are creating value but making sure to tie back that value to key objectives for the organization.
Learn how leading data providers reduce churn by using Bobsled to streamline their delivery operations. Faster onboarding, better monitoring and frictionless delivery all lead to happier customers. Get a demo today.