Airtable: Frontend as a Service
At LeaseLock, we manage a LOT of data. In our early days, we wrote basic CRUD tools to organize incoming lease applications. These tools took considerable effort to build, and our operations team didn’t exactly love them. These basic tools quickly became a bottleneck for our operations team, and we realized our product could not scale in its current form. As a result, we pivoted our product toward API integrations with our partners’ property management systems. These integrations led to a drastic increase in the data we processed each day. Consequently, we lacked the resources to simultaneously construct new data management tools and build deep integrations with our clients. We chose to focus our resources on building out the integrations, yet we knew our internal teams still needed the ability to view, structure, and manage the growing firehose of data. We had a problem.
A Discovery
No one likes building CRUD apps — especially for relational data. They are ugly, require a lot of UI work, and get outdated quickly. It simply was not feasible for our team of 4 to build and maintain a large CRUD app for dozens of tables while scaling out our system integrations. We knew we wanted a relational data editor, and we did not want to build it ourselves. Then, we discovered Airtable.
Airtable to the Rescue
We met with the operations team and demo’ed Airtable. The intuitive spreadsheet-like interface won them over immediately. The engineering team was equally excited because it provided a first-class interface for managing relational data. Through Airtable’s API, we kept the data synchronized with our SQL database; this gave our organization the best of both worlds. Additionally, it empowered teams outside of engineering to create and manage data in a way that would have been impossible otherwise.
Airtable radically improved life for our operations team. The rich feature set and flexibility allows them to work independent of engineering. Workflow tools, reporting, and integrations with other services have been built with ZERO engineering intervention. For example, operations built an entire claims product prototype using Formstack, Zapier, and Airtable.
Airtable filters, views, forms, and Zaps play a valuable part of our organization’s day to day operation. By far the most valuable of these concepts are Zaps that push information into Slack. These Slack alerts inform the team of any important events or errors. Given the real-time nature of our product, having these notifications helps us keep things running smoothly.
Furthermore, using Airtable as our frontend lets us focus on developing the LeaseLock platform. Data pipelines and integrations are of much greater value than workflow tools. Airtable enables non-technical teams to groom and organize the data while engineering focuses on architecture and revenue-driving features. Today, engineering is no longer in the business of building CRUD apps.
High Level Tech Details
The following innovations allowed us to use Airtable as a Frontend Service:
- Save button within Airtable that allowed us to create and update rows in the database in real time.
- Rate limiter that allowed us to stay within Airtable’s API throttling constraint of 5 requests per second.
- Mechanism to atomically modify data in Airtable and our SQL database.
In upcoming posts, we will go into much more detail about the evolution and inner workings of our Airtable integration.
Conclusion
We’ve benefited from Airtable as an alternative to building CRUD apps and workflow tools. With our relatively small engineering team, this has been a huge victory. Our efforts are focused on tasks that are much more important to our business. Going all in on Airtable propelled our business forward much quicker than we could have without it.