So we faced some challenges in understanding the tool and raised a couple of tickets to better understand the tool. And the improvements, I would say, would be proper documentation. And there were certain keys or maybe columns that we could not understand, which were auto-generated.

So for that kind of documentation, if that was provided to us, we would have leveraged it more, or maybe some customization around that would be the approach going forward. We would leverage that and then use it accordingly.

When processing data from certain tables with a large volume of data, we encounter significant delays. For instance, when dealing with around one million records, it typically takes three to four hours. To address this, I aim to implement performance improvements across all tables, ensuring swift processing similar to those that are currently complete within seconds. The performance issue primarily arises when we analyze the inserts and updates from the source, subsequently dropping the table. While new insertions are handled promptly, updates are processed slowly, leading to performance issues. Despite consulting our Qlik vendors, they were unable to pinpoint the exact cause of this occurrence. Consequently, I am seeking ways to optimize performance within Qlik Compose, specifically concerning updates.

We face some issues with data replication, however, Qlik helps us resolve issues.

I'd like to have access to more developer training materials. 

They need to add in more ETL features. 

It could enhance its capabilities in the realm of self-service options as currently, it is more suited for individuals with technical proficiency who can create pages using it. When it comes to end users who may lack technical expertise, they are limited to toggling between existing developments. To empower end users to make critical changes without relying heavily on technical expertise, it would be beneficial to introduce more user-friendly features for development and modification. If it could incorporate correlation analysis capabilities into its platform, especially in a user-friendly manner, it would greatly enhance the tool's overall utility and make it an even more outstanding solution. There is a room for improvement regarding stability.

The solution's setup process could be more straightforward. They should add proper documentation to understand the procedure for implementation.

They should add more functionality to NoSQL databases. There could be more customization options as well. In addition, they should improve the solution's integration with Linux systems.

Currently, Qlik's ETL and data transformation could be better. Also, the implementation can be improved because it is complicated. At first, it's really difficult for partners and clients to understand the platform and to work with it. The integration is very siloed within the IT department and IT architecture teams, which causes friction when trying to distribute it to businesses.

Additionally, they don't have the most user-friendly platform for visualizing data. Qlik was initially a data visualization tool, but it became a full data pipeline company after acquiring many companies. However, even after three or four years, they still have some challenges with data integration implementation within the clients.

Moreover, their support team could also be better. The support is also not very good.

Since it lacks in certain areas pertaining to scalability, there is a scope for improving the solution's scalability.

Qlik Compose should consider the fact that there are complex SQL statements without writing codes, documentation is not simple, SQL code is needed in order to make a transformation, and data quality functionality in the solution, and these issues need to be considered for improvement.

The solution has room for improvement in the ETL. They have an ETL, but when it comes to the monitoring portion, Qlik Compose doesn't provide a feature for monitoring. They should have some monitoring features included in it. In case of any job failures, the solution should be on an auto-fix. 

There should be the ability to trigger the jobs and schedule them. Qlik Compose can't really do that yet. If you talk about a competitor like Informatica, they do have some scheduling platforms available. 

There are basic features in Compose, but not as advanced features for monitoring and scheduling. Monitoring any scheduling features just needs to be in there in the future.

For more complex work, we are not using Qlik Compose because it cannot handle very high volumes at the moment. It needs the same batching capabilities that other ETL tools have. We can't batch the data into small chunks when transforming large amounts of data. It tries to do everything in one shot and that's where it fails.

Qlik also needs the ability to improve more complex combined SQL tables and perform more complicated transformations. I'd like to see better connectivity to some with custom coding languages like Python or Java.

It would be better if the first level of technical support were a bit more technically knowledgeable to solve the problem. I think they could also improve the injection of custom scripts. It is pretty difficult to add additional scripts. If the modeling doesn't give you what you want, and you want to change the script generated by the modeling, it is a bit more challenging than in most other products. It is very good with standard form type systems, but if you get a more complicated data paradigm, it tends to struggle with transforming that into a model.

When you replicate data, you need to transform the data into formats that are easier for querying and reporting, or data warehousing. I think the set of transformation functions can be improved. In large organizations, in particular, there are many dependencies between different paths. So to easily manage the ETL environment, you need to have visual data flow management capabilities. I believe that visual data flow management should be improved.