What is our primary use case?
I mostly use it for data insights and industry analytics. There were two use cases that were quite opposite to each other.
One was a truck manufacturing and transportation company. That company both manufactures and runs a logistics company. They use SAP Analytics Cloud for tracking their transportation, to see exactly where the consignment is, where the truck is, what the issue is with the truck, why it failed, where it failed, why it stopped, why the driver is deviating from the defined route, etc.
For the second part, an oil and gas company from the US is using it for basically monitoring their oil wells. They track how much the well is getting explored, which well is currently being used, when it can next be used, etc. So they don't continuously explore a well. They are actually making a rotation between the wells. They keep track of which well is getting explored today versus tomorrow.
How has it helped my organization?
SAP Analytics Cloud is coming up with all the algorithms in-built as a function or cloud. We leverage those for predicting different scenarios. SAP has provided a good package for predictive analytics.
We were able to convert everything we came across.
What is most valuable?
A few things are industry norms or standards. A data warehouse is available to pull data, change data types, and play with it (manipulate it). All data types don't allow easy changes. For example, if it's numbers, you cannot search with regular expressions. However, some things require exploring or finding regular expressions to show, decide, or conclude.
In the data warehouse, you're free to make changes. You can make everything text, meaning it's always a string. Even numbers appear as text, giving you the flexibility to work with all columns.
This is a regular feature of a data warehouse, but SAP Analytics Cloud also offers an out-of-the-box visualization package. This is handy for showcasing data, and everything is scalable. You can make changes and codes to see data differently. You can add different visualizations that aren't available in the package, and connect to third-party systems easily.
You have the flexibility of reading metadata. Based on the schema details, you can make necessary changes and resend data while complying with the same schema during syncing. This is useful if you are processing data. Command-line interface (CLI) mode is also available.
What needs improvement?
Some algorithms are complex and can't be processed by just tagging the data and expecting results. Perhaps SAP could add documentation that data needs to be preprocessed before processing.
For how long have I used the solution?
I have been using it for four years now.
What do I think about the stability of the solution?
There are two things to consider:
When we talk about SAP Analytics Cloud, it has two parts.
- One is hardware, which you were calling the data warehouse.
- And second one is the software with machine learning algorithms, packages, or functions.
- There's a third component for coding: either the coding IDE, the in-editor environment, or the code-level integration interface (CLI).
Both modes are available. From that perspective, it's stable.
From a development perspective, there are enough algorithms available for almost all scenarios. Where it's not available, you're free to add imports, so you're not limited.
The hardware part is largely supported, and SAP is largely using AWS, so it's okay.
What do I think about the scalability of the solution?
Regarding scalability, I'm referring to what we can do and update and whether there are limitations. So, limitations only apply to the service size you're on. Based on that, we can process data. But that's the hardware part, which SAP doesn't provide as part of the cloud offering.
On the cloud, SAP allows you to scale up the server size, and then it should work. That's what we observed from a cloud standpoint.
From a scenario-capturing standpoint, I was able to capture almost every scenario I've encountered. There was no issue, even with the complex situation at the hackathon. We simulated a load of around 2,050 and were able to make that possible in the cloud, which was a good use case.
Besides us, other companies like Accenture, IBM, Deloitte were there. We were the only small-sized company. We are small compared to other big companies participated there.
How are customer service and support?
The customer service and support are okay. The response time is faster.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
I have worked with Azure as an alternative for two of our clients. They also offer cloud analytics functionality. Their prices are lower, and you might save 40% to 60% compared to SAP Analytics Cloud.
Price is the main part. The second part is their integration with Excel, since it's their own product. Because of Power BI, they have an advantage. They can enable everything they manipulate on the cloud in Excel as well.
Although SAP also provides integration with Excel, it takes a longer route since Excel isn't an SAP product. SAP might say that Excel and Power BI licenses are subjective. Azure offers a package that includes everything, so you don't count Excel as an additional cost.
How was the initial setup?
The deployment is easy and quick. You don't have to worry much. The steps are provided, and if you follow them religiously, you're done. It doesn't take much time to enable. It's actually the client side (because of their approval protocols), like user enablement, authorizations, etc., that takes more time than SAP's part (as the document and steps are quite elaborate).
If all authorizations are in place, installation takes around a week maximum. It can be done earlier, but some things take time to enable. Last time, it took one working week to enable us, and then around a week to create the tenant and environment (with integration to Backend server). So, it was a two to three-week job.
What about the implementation team?
For deployment, typically, one person handles everything. We added a second person for support when the first person was unavailable.
One person cannot be available all the time, so we added a backup. Typically, one person should be enough, but a backup should be provisioned. For continuity, a team of two would be good enough.
SAP has out-of-the-box connectors and adapters that allow us to connect to virtually all data sources. SAP has over 350 connectors and adapters combined. So it's really easy to integrate this product with other data sources and systems that we currently use.
What was our ROI?
The ROI was faster for the oil and gas company, achieved within 18 months. For the logistics company with analytics (refer to the use case Q/A above), the ROI was achieved within 23 or 24 months. Anything within three years is considered a good investment.
What's my experience with pricing, setup cost, and licensing?
SAP pricing tends to be premium, so it is slightly higher. However, in this case, the premium charges are justifiable because it comes with out-of-the-box connectors and adapters and can still communicate with your on-premise or any other system. It creates a common interface for gaining data insights.
The only suggestion I have is regarding enabling people on the platform, which requires an individual license. This is a concern among clients because even for just viewing, a license is needed since an S-user ID or P-user ID would be created (S-users are usually direct employees, and P-users are contractual employees). Only then can that person access the cloud and the analytics. This can be a bottleneck because clients need to decide who should be granted access.
For example, in a use case where vendors were considered for access to a common solution, the solution met their expectations, but they limited access to only a few vendors. It would be beneficial if this could be adaptable for certain portions, such as viewing only. Maybe a package could be offered for that, which might help SAP expand further.
So, the individual license requirement could be there. Perhaps something for only visualizing or viewing with no changes could be offered at a lower cost or bundled. This would allow companies to enable their ecosystem partners on the platform.
What other advice do I have?
Overall, I would rate it a nine out of ten. It's a good product. I would definitely recommend using it.
SAP Analytics Cloud is typically based on machine learning algorithms, which are loosely called AI-enabled. AI has three components: machine learning, deep learning, and then AI. AI isn't widely used, but even for machine learning components, we generally call it AI.
Anything beyond machine learning components, almost all algorithms are available. If, for any specific reason, you need something not available, it can be included. So, from an AI perspective, it is very compatible and ready out of the box. All of the data insights are supported by AI algorithms.
People often refer to machine learning as AI, even though AI is typically machine learning or deep learning enabled. Both are present in SAP Analytics Cloud, so it is already AI-enabled.
Disclosure: My company has a business relationship with this vendor other than being a customer: Partner
