About the author
Peter Watt is a product manager with ChurnZero. He started his career in customer success, giving him a ground-level understanding of the problems his products are built to solve.
**
Quick summary: What REST APIs are, what they make possible, and why CS leaders should care. The examples below are drawn from ChurnZero, but the principles apply to any CS platform you’re evaluating or running today.
What is a REST API, and why should CS leaders care?
A REST API is an interface that lets your customer data move into any environment — a data warehouse, a reporting tool, or an AI model — without manually exporting a file and dropping it somewhere.
CS teams reach for a REST API alongside their platform for two reasons. First, to build a connection to a system that has no out-of-the-box integration. Most CS platforms ship with a library of native connectors, but if your team runs analytics through Databricks, there may be no pre-built option, and the REST API is how you build that bridge yourself. Second, when a native integration exists, but your business has specialized use cases it can’t accommodate. Native connectors are built for the 80% use case. For the other 20%, the REST API is where you build your own.
When you build a connector for a SaaS tool, you’re building for the 80% use case. For whatever business reason, some teams have very specific needs that can’t be configured through a normal integration. That’s when you fall back to the API and build your own.
One detail to know up front: the REST API is read-only, meaning it extracts data out of the platform rather than writing data into it. Feeding data in from an external system happens through native integrations or the HTTP API. The REST API is for getting your customer data to wherever else it needs to go.
The most common destination is a data warehouse, where teams typically run analyses in a reporting tool such as Tableau or Power BI. Some skip the warehouse and connect the REST API directly to those tools. ChurnZero offers guides that help CS teams make those connections without a developer.
How REST APIs affect the quality of AI-driven insights.
Context is what makes AI smarter. Your organization may have teams using AI tools every day who have no access to your CS platform. Pulling your customer data through the REST API into a shared data environment gives those tools what they need to produce useful output. A model operating without a customer’s health trend, recent engagement, renewal timeline, or segment produces generic answers.
Your whole organization may be using AI, but not everyone might have access to your CS platform. The best thing you can do to make AI smarter is give it context. Pull your customer data through the REST API into a shared environment, and those tools have what they need to produce useful output.
The most common integrations CS teams build, and what they solve.
The REST API is valuable for CS leaders who need to pull data and shape it in ways the platform’s native reporting doesn’t support.
Someone once asked me: ‘Can I see all my health scores for the past three years, broken down by what factors were being calculated at the time?’ All of that is available through the REST API — health score, health score factors, and calculation history. You can pull it all, connect it, and build any report you want.
You can connect your customer data directly to Excel, pulling accounts, health scores, journey data, or any combination of objects into a live sheet that refreshes automatically. For teams already working in Excel, that means a live connection without a standing export process. For teams working in an LLM environment, the same connection lets the model pull and analyze data directly.
What to ask your platform vendor, especially when AI is part of the pitch.
When you’re evaluating a CS platform and AI capabilities are part of the pitch, start with whether it has a REST API. Most enterprise platforms do. Smaller ones may not. For teams with simple data needs, that may not matter. For teams that want to connect customer data across a broader tech stack or feed it into AI tools, it matters a lot.
At the baseline, the question is: ‘Do you have a REST API?’ It’s table stakes at this point, but a lot of smaller platforms still don’t have it. The harder question is: do you have any integrations built to other systems’ APIs today? If you do, you’ll probably need something similar for your CS platform and that’s the question to ask before you’re in an evaluation, so you know what to look for.
Look at your own stack first. If your team has built custom integrations to other vendors’ APIs, you’ll likely need something similar for your CS platform, and knowing that going into an evaluation tells you what to look for.
From there, ask vendors which data objects are available via the API and whether any more are in development. Ask whether the API is read-only or read-write, and how it’s authenticated, whether through API keys, username and password, or OAuth. These questions determine whether the platform can participate in the data architecture your organization already has.
Why does this matter more as AI becomes central to customer success?
Before AI, moving data between systems required a developer to write and maintain custom integration code that connected the REST APIs of two platforms, managed data transformations, and handled edge cases.
In the old world, if you wanted to push something from a Google Doc to Jira tickets, you’d need coding knowledge to build a REST API connection between the two systems. We’re still doing that today. but instead of a developer as the intermediary, the MCP is the developer on the fly, creating that integration and sending data where it needs to go.”
See what’s possible when your customer data moves where your team does. Talk to ChurnZero.
