What needs improvement with SAP Analytics Hub?

Improvement is needed in data modeling, particularly on the Datasphere layer. SAP must also enhance AI modeling and LLM aspects. The knowledge graph and models are gradually expanding in SAP, and as people begin using AI/ML in SAP Business Data Cloud, this newer platform will mature over time. In addition to data modeling, I would like to see more automation features. SAP is already working on increasing data products, aiming for a plug-and-play mechanism tailored for various industries. Many essential data products are still unavailable and should be developed further. Moving from BW to SAP Business Data Cloud could be more seamless, similar to how industry-specifications travel. SAP is certainly aware of the necessity for improvements and is progressively enhancing the data product offerings.

SAP Business Data Cloud offers strong capabilities, but there are a few areas where it could be improved. Learning the platform can be challenging for teams new to SAP modern data ecosystem. Customization flexibility, performance optimization guidelines, third-party integration experience, as well as cost and licensing transparency still need to be addressed. Additionally, automated tools for migrating legacy systems and SAP BW HANA environments would reduce implementation effort. From a user experience perspective, I would like to see more intuitive navigation, simplified administration, and guided workflows for common tasks. New users can sometimes find it difficult to understand where specific functionality resides across BDC data space and other SAP data services. A better self-service capability for business users is also desired. On the support side, additional implementation best practices, industry-specific reference architecture, and migration plans are needed. Overall, the technology foundation is very strong, so most of the improvements I would suggest are focused on ease of use, faster onboarding, better guidance, and stronger customer enablement. Regarding SAP Business Data Cloud's AI capabilities, its governance and security are among its biggest strengths. What stands out to me is that AI initiatives are only as good as the quality and trustworthiness of the underlying data. SAP Business Data Cloud has strong data governance, with centralized control over data access, usage, and lineage. Role-based security and access management ensure users only see the data they are authorized to access, with data lineage and traceability ensuring consistent business semantics and compliance. I would like to see more AI-specific governance data dashboards and monitoring capabilities, enhanced explainability features for AI-driven recommendations, and more automated data quality assessments while using AI. Overall, I would rate SAP Business Data Cloud's governance and security for AI workloads very high, as they are among the reasons I view the platform favorably and help organizations adopt AI responsibly. Based on my experience, I believe additional improvements beyond the areas we have already discussed would include enhanced cost transparency, deeper cross-platform integration, advanced metadata and lineage visualization, proactive optimization recommendations, and DevOps and lifecycle management. Overall, SAP Business Data Cloud is already a very strong platform from a technology and architecture perspective. Most future improvements should focus on simplifying adoption, increasing automation, improving transparency, and helping customers achieve value with less implementation effort. These enhancements could move the platform closer to a perfect experience for enterprises and AI workloads.

There must be more AI included within SAP Business Data Cloud. BW modernization has been complex with SAP. If it had been simplified into more detail, that would have been good. Data governance which can be automated would be beneficial. Manually doing it each time will be a hectic job. Security-wise, I feel it is good, but data governance needs more improvement. More AI improvements are still required. It is not 100% accurate, but it needs more improvement. I would say it is 80 to 90% accurate and reliable, but still, I do not feel satisfied. It needs more improvement.

SAP Business Data Cloud is still in a phase of considerable evolution with a long roadmap. SAP has planned many releases and features, especially around accelerators for migrating and upgrading BW, which has end-of-life support in 2030, onto SAP Business Data Cloud. A strong roadmap exists for the next six to eight months. The BW journey will be tricky for organizations with heavy legacy BW footprints. Since this is a greenfield implementation, it is not easy to implement on the go. How organizations choose to move from BW on-prem to BW PCE and conduct greenfield Datasphere implementation to retire BW because of end-of-life support is going to be a critical decision, and how SAP helps organizations accelerate that journey over the next three years will be critical. BDC Delta Share had a limitation of not allowing delta only, as the delta only feature on replication flows was not available for standard ERP replication flows. This was a significant limitation that prevented organizations from migrating from BODS to replication flows. As a result, it mandates loading the entire history from source, which is not recommended and not possible. SAP is planning to overcome this limitation in the next one or two quarters with releases enabling delta only. Additionally, more accelerators are needed for how BW greenfield implementation can happen. The ABAP stack does not work, customer exits do not work, and all custom ABAP logic has to be rewritten. Accelerators for how history data can be migrated are needed. SAP has releases planned for the next two quarters around accelerating the BW migration roadmap to Datasphere. Along with that, they need to increase their scale-out capabilities on cloud, as they have limited hardware capacity on cloud. This scale-out is due in the next two or three quarters, which will help organizations speed up.

The performance of Data Product Generator to load data from BW to Data Sphere is really slow. SAP can bring up new, better ways to load data from BW to Data Sphere. We can bring in master data attributes and texts and transaction data, but there is still more improvement needed for bringing the hierarchies. SAP can definitely work on this. Implementing SAP Business Data Cloud should be taken up as a new functionality that SAP has provided. It is not replacing any of the existing system that BW or the legacy system used to do. We are going one step ahead on those. Maybe some legacy functionalities are not yet available or will not be available in SAP Business Data Cloud, but there are definitely other useful functionalities which can be leveraged and should be upgraded to. Regarding SAP Business Data Cloud's AI capabilities, the approach that SAP has taken up is the zero-data copy or zero-data sharing. The data is not moving out of SAP Business Data Cloud servers or the SAP ecosystem. The governance part is of less worry that we are not moving data, so we do not have to apply a lot of governance in both the inbound and the outbound system. It is really good.

I would prefer to see more structure in the AI use within SAP Business Data Cloud. It's still a bit bumpy, but I understand that implementing AI in any ecosystem is relatively bumpy because it's a matter of trial and error. I would prefer to see it grow more efficiently, especially the AI helper tool. There's a built-in chat feature that you can have with AI that helps give recommendations on what to do and what to improve when you're modeling on SAP Business Data Cloud or possibly doing visuals and dashboards. I would love to see that improve and give a wider scope and handle the prompts more efficiently and not be so particular. You can give it the simplest task and it should actually understand you from the first attempt. The AI features in SAP Business Data Cloud are definitely more advanced than what it used to be when there were standalone modules for everything on SAP. I'm excited to see how further it will develop because it's still a work in progress. The module is relatively new and definitely needs some tweaking, but I have confidence that SAP is very powerful and will be able to get this up and running and working just as efficiently as all of their work in progress. I'm ready to explore it.

What I dislike about SAP Business Data Cloud is that the way they position the data products is a little confusing for users because it is fundamentally different whether you're using data products that SAP delivers or customer built data products. SAP built data products are really valuable content to speed up the implementation cycle. Also, one needs to create custom data products to be able to share data from Datasphere to Databricks or Snowflake. But that is really not the full-fledged data product feature set that SAP has when they offer their own. So the data you're sharing with the other systems is just a table. That is confusing for users or potential customers because they think they can share the same objects you're using for reporting with Databricks, and you cannot. You can only share tables. If SAP was a little more open and transparent about those differences, I think that would be better. For everything related to SAP Business Data Cloud, there are still a few things that are a little quirky. I would give the entire solution an eight out of ten.

The partner ecosystem for SAP Business Data Cloud needs to be expanded as it is currently very limited with two partners, Databricks and Snowflake, but other data platforms should be added at scale. Performance is a limitation that SAP Business Data Cloud has right now, and until I know more about SAP with the Iceberg adoption, I will say more, but now performance is a limitation using the Delta capability. Regarding improvements needed for SAP Business Data Cloud, I find that there is no capability to connect SAP Business Data Cloud to AWS, so a singular improvement would be to allow myself to connect to other clouds, but also the ability to bring my own SAP servers to the platform.

SAP Business Data Cloud needs to be more reasonable in terms of costing. There is a new condition that SAP has provided to new clients desirous of purchasing SAP Business Data Cloud licensing, which is that they can only purchase it as a whole, including Datasphere and SAP Analytics Cloud. If clients desire to purchase only Datasphere or only SAP Analytics Cloud, they would not be entertained. This is one thing that could be improved. Secondly, SAP should try convincing clients to move to SAP Business Data Cloud with a more aggressive mindset so that SAP S/4HANA and SAP BW or SAP BW/4HANA become obsolete, and reliance is only on SAP Business Data Cloud for all reporting. The user experience with SAP Business Data Cloud is satisfactory; it is a very user-friendly interface. Regarding the technical aspects, I feel they need to inculcate more flexibility when it comes to using Databricks. Currently, Databricks is only being merged for the purpose of creating AI models. If Databricks could be merged for ERP purposes as well, it could benefit them significantly because petabytes of data could be processed in seconds, something that Databricks is well equipped to do. Expanding Databricks' involvement would be very helpful.

The main areas I see needing work for SAP Business Data Cloud include user onboarding of new users, more pre-built integrations with third-party systems, enhanced self-service analytic features, and additional optimization for large-scale implementations. Better documentation and training resources would be helpful for the organization to adopt the platform more quickly. The user interface of SAP Business Data Cloud could be more intuitive, particularly for non-technical users, and I would appreciate simpler navigation, user configuration, and better troubleshooting tools to improve the overall experience. Additionally, pre-built templates, integration, and guided workflows would also help to reduce deployment and management efforts. Beyond the areas already discussed, I would appreciate more AI-driven capabilities, additional pre-built industry solutions, and enhanced usability and integration for SAP Business Data Cloud. These improvements would help the organization adopt it more quickly.

Overall my experience has been positive, but I think SAP Business Data Cloud could be further improved on onboarding experience for new users, because the platform brings together multiple concepts such as data integration, governance, modeling, and analytics. There can be a learning curve for people who are new to the SAP ecosystem. I would also appreciate more guided tutorials and examples and learning resources focused on real-world business scenarios. That would help new users become productive more quickly. Another area of improvement could be simplifying the configuration and modeling processes, especially for users coming from a non-SAP background. Making these workflows more interactive would improve adoption and reduce the time required to get started. One area that could continue to evolve is AI experience. SAP Business Data Cloud already provides a strong foundation with trusted business data, but additional AI-driven recommendations, automated insights, and natural language analytics could make it even easier for business users to extract value from the data. I also think expanding and simplifying integration with a wider range of third-party platforms and data sources would be beneficial, especially for organizations operating in a hybrid environment with both SAP and non-SAP systems. Overall, the platform is very strong in data management and analytics. The improvements I would appreciate are mainly around making advanced capabilities more accessible and reducing complexity for new users.

For improvement, I would like to see more customizable dashboards, better reporting flexibility, and additional automation for data management tasks in SAP Business Data Cloud.I gave it a nine because it has improved data integration and reporting significantly. The only reason it is not a ten is that I would like to see more reporting flexibility and automation.

With SAP's AI capabilities in SAP Business Data Cloud, there are some parts integrated, but I am not convinced or impressed as much as I am with traditional data warehousing. For example, there was one component called data generator available in SAP Business Data Cloud. It was transforming a previously built data model in BW to the DataSphere model. However, we have seen some disturbances where the data model built on custom functional modules needs human dependency. It was not transforming exactly as much as our requirement. That is one point. The second point is about the Insight app; I am not that happy about this. It can be improved. For the Insight apps, they need to be shared through other reporting platforms because of client requirements. One of my clients was from the manufacturing sector and wanted to try the Insight app. They wanted some reports in SAC and some reports in Power BI. However, the Insight app is not available; we cannot share this Insight app to Power BI. That was the issue we faced. This is a limitation. In SAP Business Data Cloud, to help different AI assistants stay in sync and share the same business rules so they do not give conflicting information, I feel SAP still builds all these AI capabilities into a closed system. If you compare other data engineering stacks, they are openly integrating and partnering with other platforms. They are much more advanced and much more ahead of their time in comparison to SAP Business Data Cloud. SAP Business Data Cloud does support AI or ML enabled with new use cases in our organization. There are some apps we want to develop, and it is not only limited to SAP Business Data Cloud. We are trying to integrate Joule capabilities into SAP Business Data Cloud. For example, if we are doing a greenfield implementation, on top of these tables, we have to create the CDS view for optimized extraction. In that case, we are trying to do this CDS extraction and CDS code writing using Joule AI automation. That is something SAP could directly integrate into SAP Business Data Cloud. It will save us time. Second, they have given the product generator, so remodeling is a bit easy. The third point is about reporting. The Insight app is not something I am happy with overall. The Insight app concept was not giving end-to-end functionality, and there is a limited scope of customization into that pre-built Insight app. SAP can work on improving this.

SAP Business Data Cloud could be improved by publishing broad guidelines on how to handle it correctly, perhaps specifying certain things not to do. For instance, my current structure requires extracting, transforming, and loading data, but Business Data Cloud operates more in BW modeling in HANA mode. You still need your SQL joins, but you only need to persist data when necessary. If it can introduce agents to accelerate data modeling or aid development, that would provide extra flexibility. For example, if I need to make customizations, I could define specifications, and an agent could develop that on BTP and integrate it with my BDC, giving me additional flexibility and a better time to value from the development cycle while adhering to SAP's recommended coding standards. I would not honestly give any product a perfect rating. However, one area where Snowflake has gained traction is dynamic workload management. I am uncertain if SAP Business Data Cloud provides that feature, but if I have certain reporting demands and the system is at its limit, jobs can fail. It should offer automated workload management, and while SAP can charge for it, such a feature would prevent higher workloads from causing system failures and disruptions.

SAP Business Data Cloud can improve by reducing the licensing cost because each level requires clients or partners to pay different licenses based on its functionalities. For instance, if we purchase SAP Business Data Cloud only for SAC, we pay the SAC licensing, and if we also purchase Data Sphere, we then pay for Data Sphere licensing, Databricks licensing, and BW bridge licensing. At each step, partners or organizations requiring SAP Business Data Cloud pay licensing for every tool. If we can reduce the licensing cost, it may help small or medium organizations to implement SAP Business Data Cloud, making it more popular globally. My experience with pricing, setup cost, and licensing of SAP Business Data Cloud is that it is very high compared to implementation costs. If we could reduce this cost based on capacity units, then small and medium organizations could also avail themselves of this opportunity.

SAP Business Data Cloud can be improved as some of the features from BWPC are still not available in the Data Product Generator. Including features such as multi-provider or legacy BW object access could be helpful. Also, if connections to third-party tools such as Power BI and Tableau can improve, it would enhance customer interaction. Additionally, if SAP can come up with innovative options to reduce licensing costs, many customers would incline toward SAP Business Data Cloud since some are currently moving towards non-SAP tools such as Fabric, Databricks, or BigQuery due to cost constraints. The performance when utilizing the zero-copy data share via BDC Connect could be improved, as it would be very useful for most customers who start using it. Right now, some customers face issues because of BDC Connect. I have already mentioned all of the issues I faced, so I don't have additional improvements to add.

I think the areas of SAP Analytics Cloud that could be improved or enhanced are mentioned by users or companies, and they would definitely mention the price in terms of planning licensing. From a consulting point of view, I don't see any major weaknesses. It is reasonably well-developed, and the perfect strength is that innovations are coming on a quarterly basis. It is slightly behind other tools in terms of the number of graphical outputs. Some tools have it faster, but every increment of innovations contains new features. On the other hand, it is perfectly fitting to the SAP environment, with seamless integration to all SAP modules, not only ERP but also SuccessFactors and other tools from the SAP family, for example, integrated business planning for the production planning. This is really a highlight compared to other tools.

The search function in SAP Analytics Hub is not effective. When speaking about integration with various analytic sources, for example, if I wanted to send the data from SAP BW or B4HANA to a data lake, we used to earlier use the Open Data Hub or Analytics Hub. Nowadays, we are using the Data Sphere integrators available with Amazon, Azure, and Google Cloud. The search capability could be better, and SAP needs to come out with a clear roadmap for the solution going ahead, as they have not yet provided that.

The platform could easily integrate with other ERP data solutions. Additionally, it should be faster compared to competitors in the cloud space.

SAP does not have any capability of self-service analytics. It has a GUI, which has nothing to do with end-user empowerment. It should work more on integration. While SAP Analytics Hub doesn't directly allow manual metric creation from uploaded Excel files, it seamlessly connects to various data sources, including Excel files, to create comprehensive analytics and visualizations.

Faster implementation of new functions for the new product is an area of improvement in SAP Analytics Hub.

Pricing and support are integrated into the system. However, the support process can be challenging, particularly when navigating the SAP support site, which lacks user-friendliness and ease of access. A good level of understanding is needed to use the system efficiently. Users who are unfamiliar with the topic area may find this difficult.

The technical support for Analytics Hub is limited and could be improved. It is very difficult to import data from other products, such as Power BI. In general, it doesn't integrate well with non-SAP products.

There is a steep learning curve with SAP Analytic Cloud for others from Non-SAP world, compared to other cloud solutions like Azure, GCP, AWS etc. Access to SAP training labs are a challenge compared to other vendors and the learning content can be structured better. Now, the content is spread across developer forum. Practice labs can be improved. Specific learning paths are not available for some of their products like SAP Data Intelligence which needs to be integrated with SAP Analytics Cloud and there are no hands on Labs to practice SAP DI on Access. The learning interfaces on Developer forum can be more user-friendly experience. learning experiences on GCP & Azure are a breeze.

The analytics part of SAP has not been that good up until now. They're still in the infant stage and development is still going on. SAP has some cloud analytics and they're trying to configure it in Lumira, but it's not that good at this point.