What is our primary use case?
Creating analytical models that we put into production: Everything ranging from pricing to just-in-time inventory management.
We have had multiple models go into production. We are at around roughly 10 models right now. We were able to quickly transform and move existing models into the SPSS environment, so we saw increases in accuracy resulting from this. Therefore, we are running faster and more accurately.
This is batch. We are using models for safety and to predict what drivers are likely to leave (i.e., just-in-time inventory management), so grows it across the enterprise.
We're using a public Azure cloud. We are not deploying apps, but we are doing the analytics. We are pulling the data in with it, then we are writing the tables.
It has performed as it should. I have not had any issues.
How has it helped my organization?
We are creating models and putting them into production much faster than we would if we had gone with a strictly, code-based solution, like R or Python. In the time it takes to write the code to build one model, I am building three models inside SPSS.
What is most valuable?
- The ability to quickly prototype.
- The integration into all the existing environments.
- The ability to not have to manage a lot of code.
What needs improvement?
I would like better integration into the Weather Company solution. I have raised a couple of concerns about this integration and having more time series capabilities.
What do I think about the stability of the solution?
It works fine. I have not had any stability issues; it is always up.
What do I think about the scalability of the solution?
It scales. I have not run into any challenges where it will not perform.
How are customer service and technical support?
Technical support is great - 90% of the time.
Which solution did I use previously and why did I switch?
The organization did not have a solution before this one. I was familiar with SPSS having worked there. I knew its capabilities and got them involved on the front-end.
How was the initial setup?
The initial setup was straightforward. Though, I had done it before.
What was our ROI?
I have never done studies on the time savings. Based off the ability to build codes quicker, then put them into production because we have collaboration employment services which is another analytic solution from IBM, so we are able to productionalize the models and manage the models from this environment. Altogether, this saves us a lot of time versus if we want a programmatic solution and had to have developers write C# and Java around it. Overall, it is a huge increase to time savings.
Which other solutions did I evaluate?
I looked at Microsoft and Alpine Data. I also considered SaaS.
I chose IBM SPSS because of their experience with the solution, what they brought to bear, and their relationships.
- They are established.
- Having worked there, I knew the tool, I had used it in prior roles.
- The cost models: I prefer to own a solution versus leasing, much like a SaaS solution. This was one of the things that stood out.
- I know the product managers for SPSS and where they were heading from a roadmap solution, and it is very much aligned with what I was trying to do.
It was this altogether, as well as the price.
What other advice do I have?
Take your time and do some PoCs with this solution and other solutions. At the end of the day, you will be highly impressed with SPSS capabilities and the capability to get models into production. You should take a hard look at SPSS.
Most important criteria when selecting a vendor:
- The vendor's willingness to invest in the relationship
- Vendor's experience
- Product's stability
- Bringing the enterprise solution to bear.
There are a lot of vendors out there that have been around for three or four years, what I would consider startups. Then you have enterprise solutions, which have been around for 20 or 30 years.