How to successfully run 400 models every hour with a 100% uptime
100% uptime

Makes it possible to run over 400 models every hour without lag

Unmatched shipping speed

Less than two months from customer request to shipped feature

Straightforward pricing plans

Innovative tools shouldn't cost a fortune

An arrow symbol used for links
Back to all customers

The leading legal tech company in the Nordics, making law easy for everyone.

Stockholm, SE
Professional services
An arrow symbol used for links
Back to all customers

How to successfully run 400 models every hour with a 100% uptime

I. Introduction: challenging the status quo

Almost a decade ago, Lexly challenged the status quo in the legal industry, building Europe's first platform for digital law. Their vision, both then and now, is to make legal security accessible to everyone. They offer customer-focused legal solutions in collaboration with insurance companies, banks, and a vast network of lawyers.

Fast forward to today, Lexly stands as the leading legal tech company in the Nordics.

We sat down to chat with the one-man band, Henri Jäderberg, who leads Lexly's data and analytics operations. Jäderberg plays a crucial part in identifying optimization opportunities, whether lead conversion or effective budget allocation. His primary focus is tracking and measuring performance across all four markets - from unit economics to marketing automation results.

II. Challenges: running fast as a one-man band

Jäderberg strongly focuses on details, real-time, and speed, leading him to run over 400 models per hour to deliver accurate and timely insights to stakeholders, including Lexly's founders. He states, "You never know what you need until you need it." With the high frequency of continuously running jobs, tool stability becomes crucial, especially when you're only one person.

"The primary issue with our previous analytics engineering platform was its slow speed"

They endured prolonged waits for jobs to run successfully and experienced significant lag, hindering smooth operations. The constant delays due to the workload hampered productivity and made working at the pace typically expected in an SQL client environments challenging.

III. Solution: scale at speed with no downtime

Jäderberg's manager, having discovered Paradime, suggested that Jäderberg should try it. After a week of comparing the two platforms, they realized Paradime was a better fit for their specific needs. For example, the 99.9% uptime guarantee addressed some of their biggest pain points.

In his opinion, one of the most critical factors when testing new tools is how easy it is to get up to speed. Jäderberg emphasizes the importance of user-friendliness and that nobody wants to spend days setting up a new platform only to find it to be overly complicated to use. With Paradime, he successfully ran jobs within a few hours of implementation.

Features, such as enhanced preview functionality, are crucial for him to work efficiently, helping him check and test anything during the development phase and aligning with his preferred work style.

At Paradime, we build our platform together with our users, which Jäderberg states as an important success factor. We view our users as the actual product experts, and even the simplest feature requests can significantly enhance the overall user experience. Jäderberg himself, for example, suggested a feature that allows users to copy the model name directly from the file structure by simply right-clicking. Paradime implemented this feature in less than two months.

This particular feature has greatly accelerated his workflow, especially when dealing with lengthy model names. Not every new feature needs to be a LinkedIn announcement for buzz.

IV. Impact: happy stakeholders, happy life

  • Straight-forward pricing plans: With the number of jobs Jäderberg runs daily, the cost could potentially skyrocket. He says it's hard to beat Paradime's "no-fuss" pricing plans.
  • 100% uptime and no lagging: Paradime's consistent 100% uptime over the last six months has made it possible to run over 400 models every hour without any lag.
  • Shipping speed: Two months from customer request to shipped feature is the kind of speed both smaller and larger teams need to stay on top of their everyday work.

V. Conclusion

As a one-man band, Jäderberg highlights how crucial it is that the tools he uses are stable and continuously improve to keep up with the speed he needs to work at to bring real value to Lexly.

At Paradime, we will always go above and beyond for any of our partners, whether a 50-person-strong team or a one-seat client. Great ideas and feedback can come from anywhere, at any time.

Interested to learn more?
Try out the free 14-days trial

More case studies