Kanso Software boosts MongoDB efficiency by 90%, saves $36K per developer

About Kanso Software

Kanso Software is a US-based housing software provider specializing in supporting Native American housing authorities. Serving over 200 housing authorities, Kanso handles intricate data problems with tailored solutions offered through SaaS and PaaS platforms.

How Kanso Software uses MongoDB

Kanso Software offers a SaaS (sometimes PaaS) console, where each customer gets a copy of the monolith deployed, which is hosted in AWS ECS as a Fargate instance built on a MEAN stack, and on their own MongoDB production cluster hosted in Atlas, averaging 45-60 collections. Kanso Software’s product has been evolving to handle larger, more sophisticated clients by introducing significantly complex and robust features. They’ve been actively modernizing the platform and enhancing its architecture to ensure it remains a powerful, scalable solution that meets the growing needs of their users.

Supporting customers are four diverse teams that leverage MongoDB, including software developers, DataOps engineers, QA engineers, and an Innovations manager. The teams are mostly composed of junior to midlevels who have little or no previous experience working with MongoDB. Without Studio 3T they would have been limited in how they are able to efficiently and quickly perform their tasks and provide their customers the quality and speedy service they require.

What was Kanso Software up against?

The initial product that Kanso Software built was meant to replace a legacy Windows-based desktop application that had been developed for over 20 years with features being added continuously. It evolved organically and developed a rich functionality over time, which Kanso Software has reimagined using more modern technology. The product is constantly evolving to meet modern standards and align with the forward-looking vision of its users.

Here are the key challenges Kanso Software was faced with and how they used Studio 3T to ensure they can enable a great customer experience, meet deadlines, save precious development time and cut costs on onboarding and training.

Challenge 1: Working with highly nested data

Kanso Software’s documents are highly nested, with some properties containing multi-dimensional arrays. Working with this highly nested data is challenging. It’s complicated and time-consuming to leverage the data in a useful way, and requires Kanso Software to build highly creative aggregations. 

Solution: Aggregation Editor, Visual Query Builder

With Studio 3T’s Aggregation Editor, Kanso Software’s developers can view the data entering a stage and immediately see the output, making it easy to debug and refine each step in the pipeline.

The aggregation framework allows them to build very complex data pipelines, to query that data, reshape it and get it into a state that’s usable. The visual nature of Studio 3T enables Kanso Software to quickly and easily query and find exactly what they need to, no matter how nested their data is. A process that would have taken hours now takes only 10 minutes. 

Challenge 2: Moving production data

Kanso Software has over 200 production databases, with every database having around 65 collections, some of them with a very large list of logs. The Kanso Software team struggled with copying non-contiguous lists of collections between databases, clusters or environments. 

In the past, when they needed to debug anything with a production dataset, they had to copy the data set to an internal database on an internal cluster in order to avoid affecting production. This process required manual scripting, taking a developer 10 minutes around 20 times a day per developer/DataOps.

Solution: Connection Tree

Now, with Studio 3T’s copy & paste capability, this process takes only one minute – 90% less time and effort, which saves Kanso Software more than three hours a day per developer.

Not only that, but if there happen to be any copy issues, Studio 3T clearly shows the issues and what the problem was, so the team can simply grab that one collection and drop it over again.

When the team needs to restore a cluster or database in MongoDB on an Atlas cluster and restore an entire huge cluster it would take 2-4 hours, but with Studio 3T they accomplish that in only 10 minutes using Collection History.

Challenge 3: Lack of MongoDB skills/experience

Much of Kanso Software’s team doesn’t have much or any pre-existing experience with MongoDB, which would require them to spend valuable time learning to build queries in the command line, creating aggregations, moving data and any other task they need to perform. In addition, any developers joining the team would require extensive and costly MongoDB training and a lengthy onboarding process.

Solution: Visual Query Builder, Aggregation Editor

The Kanso Software team relies on Studio 3T’s intuitive drag-and drop Visual Query Builder to easily and quickly build their queries. They also use the Aggregation Editor to create complex aggregations that would almost be impossible for them otherwise. 

Studio 3T empowers the team to perform advanced MongoDB tasks without prior experience,  This reduces training time, and minimizes human error when creating queries and aggregations.

Challenge 4: Identifying and fixing data issues

The Kanso Software team is required to make many manual data fixes on a daily basis when data gets imported incorrectly or when customers make mistakes. These manual processes require valuable development resources spent on dealing with many menial tasks.

In the past, the Kanso Software team would write scripts in Typescript, which would help identify and address specific issues with their data. The scripts would run on all databases and clusters, and were designed to surface issues by running queries that gather info on data irregularities, like missing data. Other scripts were then used to modify this data.This was highly time-consuming, from writing to executing. 

Solution: IntelliShell, Team Sharing

Now with Studio 3T’s IntelliShell, Kanso Software can re-use scripts and adjust parameters and queries much faster, and the many daily data fixes they need to do that would take hours are done in a matter of minutes.

Additionally, in the past each developer would save the scripts in a repository and use them as a template for these kinds of tasks, not only for individual use but also in order to share with colleagues. Now with Studio sT3’ Team Sharing function, teams can re-use and share scripts, improving collaboration within the team, simplifying the workflow and resolving issues at speed.

Challenge 5: Converting customer data

Kanso Software’s DataOps team is responsible for converting their customers’ data: Collecting data in the discovery phase, evaluating it and then building the process to transform it into what their platform needs. This process takes a few months to complete, with an expectation from customers that this should not exceed three months.

Solution: Export Wizard, Import Wizard, Collection History

Kanso Software extensively uses Studio 3T’s Export Wizard and Import Wizard, as well as the Collection History feature to quickly and easily move and convert data from their proprietary legacy system and/or external systems into their MongoDB databases. This enabled them to iterate very quickly and convert all relevant data within three months instead of six months.

Studio 3T has also opened up additional capabilities, such as importing partial datasets, giving the DataOps team the ability to import a portion of data related to specific entities, such as waiting lists from legacy datasets ,which previously could only be imported wholly.

Without these capabilities, Kanso Software would spend double the time on this process, which translates into not only a much slower service, but also the risk of losing customers who may find the process not fitting their tight timeframe requirements. Without this capability Kanso Software would be at risk of losing $40-80K per any deal that would fall through due to not being able to meet project deadlines. In addition, data conversions done using Studio 3T have been more accurate and streamlined than before.