IBM SPSS Modeler Reviews

We are primarily interested in the supply chain data analytics, focusing mainly on procurement. We believe that there is a lot of value in spend analytics because of the following:

• Understanding and identifying opportunities.
• Reduced prices.
• Ability to better do negotiations.
• Understanding better the categories. 

We are interested in finding the right model in order to do data mining correctly. We want to learn and understand which models are best for us, then know in which cases to use them. 

Trial/evaluations only.

It is pretty scalable.

Based on several discussions that we have had with our local representative, the initial setup should be quite short, a few weeks, or two or three weeks for the PoC. We expect him to transfer the data to us, allowing our internal analysts to do the analysis. 

We have a local representative in Israel who specializes in SPSS. He will help us do the PoC, allowing us to understand if we will pursue this process.

We have the impression that he is an expert in this area. We expect him to help and guide us through the process.

We are also interested in Watson Analytics. There is an issue that we are trying to understand because we are in a different industry. We find it quite challenging to transfer the data to the cloud. Therefore, we want to understand if it is possible to do it on-premise. We are trying to investigate this issue.

I use the product for my research purposes in running analysis on data.  

The most valuable feature for me is just being able to run statistical analysis on my data.  

I actually think it is a great product. Maybe there could be some enhancement with more extensive help built into the interface. This could help end-users to understand the features as well as how to use them. Apart from that, it is a great product.  

We have been using IBM SPSS (Statistical Package for the Social Sciences) for about a year-and-a-half.  

The initial installation is straightforward.  

I did not have to use an integrator or consultant for deployment. I was able to do it by myself.  

I would recommend IBM SPSS Modeler for people who need to do statistical analysis of this type.  

On a scale from one to ten (where one is the worst and ten is the best), I would rate this product as an eight-out-of-ten. It does the job.  

On-premises

I use it for research. It is able to do the basics but does not offer much variation. There are not a lot of options.

It is easy to use and robust.

When I do clustering, I want to try a different stream, but currently the only thing that I can really pick is averaging. There are no other options. I would like to see more options added and some modeling things.

The stability is good.

The scalability is good.

JMP, because my team is moving toward SAS. I don't quite like SAS because of the user interface, but now they have JMP which is graphical and I like it.

I would recommend SPSS to someone who has just started trying to run a lot of modeling, it's a good starting point. It is very easy to use and will do the basics.

It does what it needs to do, but it's simple. If I want to modify things, then maybe I need to find something else.

The software is robust with advance statistical tools in hand from time series analysis to logistic regression, it can be used by banks for fraud detection, by convenience stores for market basket analysis, for cluster analysis on customer segmentation.

The software is quite expensive and IBM is currently marketing its other digital dashboard tools such as IBM cognos, now we aren't sure on the plans of IBM integrating these two softwares.

It's a great software for advanced users, users should have a background in statistics, sql server and strategic planning, preferably the software is for strategic planning offices and business intelligence units.

I used SPSS for statistical hypothesis analysis and it performed well. 

It helped me in that I didn't need to write them by hand, and I could get a result in one or two minutes. That helped me a lot. 

I would like see more programming languages added, like MATLAB. That would be better.

The instillation was easy.

The stability was good.

It was scalable.

 I also use SAS, but SPSS is easier than SAS, and I enjoy it.

I would advise my colleagues to use SPSS, depending on the work that they want to do. Though for complex issues I might advise them to use better software. 

• Automated data preparation
• Some basic form of feature engineering for classification models. This really quickens the model development process.
• Automated modelling, classification, or clustering are very useful as well.

Automated reporting data preparation: A lot of jobs, which were stuck in Excel due to huge numbers of rows, are now tackled pretty quickly.

• Formula writing is not straightforward for an Excel user. Totally new set of functions, which takes time to learn and teach.
• Automating procedures: Writing macros is not easy and very hard to learn.
• Unfortunately, it’s not integrated with Qlik, Tableau, and Power Bi.
• Expensive to deploy solutions. You need to buy an extra deployment unit.

I have been a user since 2010, when it was named Clementine, and worked with versions 15, 17, and 18.

This tool has a lot of bugs. The canvas gets crippled when you put on a lot of icons. You have to save, close, and open the stream from scratch.

When there is a change in regional settings, menus and so on don’t pop up, probably due to software developed with Java.

Server installation was too hard and ineffective. You need at least four to five concurrent users with huge data to benefit from a server, but the salespeople never tell you that and just try to oversell.

Really bad local support. It took a long time for support to install a server.