It is currently being implemented to replace our organization-wide data warehouse and BI reporting platform. We've initially completed the data migration, and we are starting the build out of all of our organization reports as Sisense dashboards. Secondarily, we are utilizing their API capabilities to serve data to new websites we've recently launched.
Their technical support has no option for immediate phone support in the event of a disaster or emergency. Their support staff is mainly located in Israel, so a ticketing system is relied on no matter what the urgency.
Their tech support also doesn't easily remote into your instance if you're hosting yourself to be able to quickly diagnose issues.
We've still been experiencing many random issues with Sisense like corrupt files and unexplained failures in production which seems very buggy.
Anytime deeply advanced SQL is used in the ETL section (Elasticube Manager) Sisense kind of falls apart. It needs any complex queries to be broken down into multiple, separate queries at the simplest denominator.
Sisense - in POC alongside Birst and Qlik, Sisense ran circles around Birst and Qlik and made them look stupid by comparison
Birst - ETL was super slow and complex AND requires you to learn a proprietary scripting language Qlik - No out of the box API access to your data and also requires learning a proprietary scripting language Pentaho - huge and bloated architecture with way too many moving parts
Domo - Final winner we ended up replacing Sisense with. Domo is much more stable and does everything Sisense did and didn't do for us.
Sisense has many great connectors out of the box, but when it comes to connecting to REST API's as data sources, there is no way to use an actual scripting language to generate hashes or signatures with timestamps where that is needed for secure REST API access. This was a big headache for us requiring a pretty convoluted work around.
Though it is possible to share dashboards and even collaborate on design and filtering, it's not intuitively clear what changes are limited to what you're doing and what changes will be transferred to who you're sharing the dashboard with. It ends up requiring a guess and check cycle with lots of communication to get a collaborative dashboard correct.