Case Study: How Health in Tech Maintains Databases and Security Compliance with Atlas

"Everybody should be using Atlas. If anyone is still trying to manage their database objects in some more traditional way, they really need to get on board and start using Atlas."

– Kevin Hochhalter, VP of IT Operations, Health in Tech

Company Background

Health in Tech (Nasdaq: HIT) is an AI-driven insurtech company that automates and streamlines self-funded health insurance plans for small and medium-sized businesses. Through integrated platforms like its eDIYBS quoting engine and HI Card claims system, Health in Tech reduces administrative friction across underwriting, risk management, and network pricing workflows, letting brokers, carriers, and third-party administrators design, price, and bind customized health plans within minutes.

Managing Migrations Manually

Before Atlas, Health in Tech managed schema changes the way many teams still do: by hand. Migrations were scripted and applied manually, following a traditional DBA workflow that dated back to the mid-90s. That kind of process can work well enough at a small scale, but it depends on every change, environment, and dependency being tracked correctly across the board. That gets harder to guarantee as the number of environments and the pace of change grow.

Eventually, the gaps that model leaves open showed up in practice with a batch of changes that had to be rolled back due to tracking errors, causing a full 20-minute outage. Separately, a database constraint that was present in other environments failed to propagate to production, corrupting data and costing hours of manual work to track down and fix. Underlying both incidents was the same root problem: manual processes make it extremely difficult to see how schemas differ across environments, and to manage those differences as things scale.

Looking for a Terraform for Databases

When Kevin Hochhalter, VP of IT Operations, took over database management at Health in Tech, he brought a clear idea of what a fix should look like. The team already trusted Terraform and Ansible to manage its compute infrastructure declaratively, and Kevin wanted the same model for its databases: describe the desired state, and let the tool calculate how to get there.

That search led the team to Atlas roughly two years ago, and they began rolling it out to manage schemas across their growing set of environments.

Today, Health in Tech runs a more conservative but far more reliable process: developers submit tickets describing the change they need, the database team validates the request, and Atlas applies it through a CI/ CD pipeline that enforces structural integrity before anything reaches production.

Declarative Migrations and Built-In Guardrails

For Kevin, the core value of Atlas comes down to the same declarative model that won the team over to Terraform in the first place. "Where Atlas really shines is in its declarative migrations. It gives me the ability to examine the current state of the infrastructure and calculate how I can move to the desired state. It’s brilliant."

Instead of tracking by memory whether a change has shipped to a given environment, the team gives Atlas a schema definition and trusts it to compute the exact SQL required to get there, idempotently, every time. That confidence extends to release day. "We don't spend 45 minutes backing up the database before we do a commercial release because Atlas has earned our trust," Kevin said. Atlas's dry-run capability lets the team preview exactly what a migration will do before it runs, removing the uncertainty that made those backups necessary.

Health in Tech also uses custom linting policies to keep guardrails in place around that trust. Policies block the accidental dropping of tables or schemas outright, and exclude specific partitions, such as those managed by pg_partman, from Atlas's management entirely, so the tool never touches objects it shouldn't.

Meeting Audit Requirements With Atlas Cloud

As a company operating in a heavily regulated industry, compliance isn't optional, and Atlas Cloud has become the team's source of truth for proving it. "There are many compliance measures that I have to keep track of," Kevin said. "Atlas Cloud is a great resource. It's always available and up-to-date. It satisfies my compliance requirements and I absolutely love that."

Atlas Cloud's record of who applied which migration, to which database, and what SQL was actually generated gives Kevin the evidence he needs for security audits on demand. "I had to provide evidence about how we manage these things, and I could easily go into Atlas Cloud and pull them out," he said. Atlas Cloud also serves as the team's central hub for Entity Relationship Diagrams (ERDs), giving everyone a current, shared picture of the schema without anyone maintaining diagrams by hand.

The Outcome

Two years in, Atlas has become the team's default answer for schema management. "Atlas is our go-to schema management product. It has been integral in managing database schemas while expanding our products and environments," Kevin said. What used to be a human-led process prone to outages and silent data integrity failures is now a controlled, auditable pipeline:

• Fewer late nights fighting fires: With outages and data corruption no longer traced back to tracking errors, Kevin's team spends its time on database strategy instead of post-incident cleanup.
• Audits that take hours, not weeks: What used to mean digging through logs and memory to reconstruct a change history now means pulling a report from Atlas Cloud, freeing Kevin up ahead of audit season.
• Changes don’t go rogue: With destructive changes blocked automatically, the team can move faster on routine schema requests without a second set of human eyes on every single one.
• Capital freed up for the business: Cutting the pre-release backup ritual isn't just a time save, it's one less recurring cost standing between the team and shipping commercial releases on schedule.

Getting Started

If your team is still managing schema changes through manual scripts or tribal knowledge, the risks Health in Tech faced (such as unpredictable outages, silent data integrity failures, and audit evidence nobody can easily produce) tend to compound as the system grows.

If you already trust a declarative model like Terraform for your infrastructure and want the same guarantees for your databases, or you operate in a regulated industry that needs a clear, auditable record of every schema change, Atlas could be the solution for you.