Dundas BI vs. IBM Cognos Analytics

For all the scenarios I have so far worked on or I am currently working on, Dundas BI has proved to be more than adequate and apt to handle all of those. It is a very easy-to-use tool with quick shortcuts enabling you to prepare ad-hoc reports or dashboards in a matter of minutes.

Incentivized

Our first and most basic scenario was to automatize the creation and publication/sharing of business reports that used to be created manually by employees throughout our organization. Using Cognos for that first use case worked well. The advanced analytics functionalities we used on the aggregated data were also as expected. However, the user interface is not always intuitive, and there is a steep learning curve, so I would recommend Cognos only to large organizations where there is a need for the most advanced functionalities (AI analysis/prediction).

Incentivized

• Project organization from Development to Production, you get a production and development license but I think the best way to do it is with DEV and Prod project in the Production box. Use the development box for testing updates and really crazy things. With the Dev and Prod projects on the same box, you just publish from Dev to Prod and you are done. Users only have access to the Prod projects so no one can mess up what you are working on.
• Security - If you have a hierarchy (subsidiaries, divisions, department, teams) and you want each group to see only their data, then Security hierarchies are for you!
• Dependent filters! What's this you ask? Here is an example of how it can be used, in your company you have departments and who works for what department is in your database. You make a dashboard that has a department filter (only show these departments), a managers filter, and employee filter. Not every manager or employee is in multiple departments usually only one. With dependent filters you can say that the manager and employee filter are dependent on what is selected in the departments filter so when you go to filter them they only show the managers or employees that are part of that department, and you can even it do so employees are not only dependent on department but on manager as well. Then it gets even better as it can be done in reverse as well so when you select a manager then go to the department it only shows the departments he works for (there are better situations where this is more useful).
• It is scriptable! From calculate columns, null replacements, button actions, load actions, hover over events there a way to do what you want.
• They are constantly improving and listens to your suggestions.

Incentivized

• We use the tool for data modeling as it helps in predictive data analysis for complex data, which is very similar to real-life scenarios.
• Options of customizing & scheduling reports as per our requirements basis.
• Has mobile application which works seamless.

Incentivized

• Not too many cons for how we use the application. It really is easy and powerful. Very powerful.
• Licensing is one thing that could be looked into. It is simple, but a little confusing. For example, if I get a license today, but a new release comes out tomorrow, it seems that the license doesn't work with the new release. Maybe that is by design, but it would be nice to clearly understand.

• Sometimes there might be performance issues when dealing with large and complex data.
• Although IBM provides full documentation but sometimes it's difficult to find answers to questions and connect with their customer support.
• It relies on external tools for data cleaning, transformation and also for some integration tasks. It can also improve on providing wider range of data sources for integration.

Incentivized

For an existing solution, renewing licenses does provide a good return on investment. Additionally, while rolling out scorecards and dashboards with little adhoc capabilities, to end users, cognos is very easily scalable. It also allows to create a solution that has a mix of OLAP and relational data-sources, which is a limitation with other tools. Synchronizing with existing security setup is easy too.

We are still in the implementation phase, but so far we are finding it to be easy to use and learn. The eLearning courses that they have made available for free, as well as User Forums and other training videos have made even difficult concepts easier to understand.

We have a strong user base (3500 users) that are highly utilizing this tool. Basic users are able to consume content within the applied security model. We have a set of advanced users that really push the limits of Cognos with Report and Query Studio. These users have created a lot of personal content and stored it in 'My Reports'. Users enjoy this flexibility.

Reports can typically be viewed through any browser that can access the server, so the availability is ultimately up to what the company utilizing it is comfortable with allowing, though report development tends to be more picky about browsers and settings as mentioned above. It also has an optional iPad app and general mobile browsing support, but dashboards lack the mobile compatibility. What keeps it from getting a higher score is the desktop tools that are vital to the development process. The compatibility with only Windows when the server has a wide range of compatibility can be a real sore point for a company that outfits its employees exclusively with Mac or Linux machines. Of course, if they are planning on outsourcing the development anyways, it's a rather moot point

Overall no major complaints but it doesn't handle DMR (Dimensionally Modeled for Relational) very well. DMR modelling is a capability that IBM Cognos Framework Manager provides allowing you to specify dimensional information for relational metadata and allows for OLAP-style queries. However, the capability is not very efficient and, for example, if I'm using only 2 columns on a 20-column model, the software is not smart enough to exclude 18 columns and the query side gets progressively larger and larger until it's effectively unusable.

We have bi-weekly calls with our Success Manager, as well as access to support as needed. Any question that I have had, multiple people have been willing and able to jump on a call to talk me through it, or send an email with the solution

Why is their web application not working as fast as you think it should? They never know, and it is always a a bunch of shots in the dark to find out. Trying to download software from them is like trying to find a book at the library before computers were invented.

Incentivized

Onsite training provided by IBM Cognos was effective and as expected. They did not perform training with our data which was a bit difficult for our end-users.

The online courses they offer are thorough and presented in such a way that someone who isn't already familiar with the general design methodologies used in this field will be capable of making a good design. The training environments are provided as a fully self contained virtual machine with everything needed already to create the environments. We've had some persisting issues with the environments becoming unavailable, but support has been responsive when these issues arise and straightening them out for us

Providing that you have the client's needs well known, creating BI content using Dundas BI is a walk in the park!

Make sure that any custom tables that you have, are built into your metadata packages. You can still access them via SQL queries in Cognos, but it is much easier to have them as a part of the available metadata packages.

Per dollar spent, it offers the widest range of features of the tools that we evaluated. It offers lots of options for how to configure your environment, though they are not always intuitive to figure out. Having an ETL layer was a must have for us, as well as the ability to host to secure HIPAA compliance. It is not a replacement for ad hoc reporting, but does a great job of creating parameterized reports and dashboards that look great.

Incentivized

In the past Management had used Excel and Workiva capabilities to create the reporting dashboards that were being used to make decisions. Since switching to IBM Cognos Analytics the Company has been much more efficient and decision making has been streamlined. IBM Cognos Analytics was chosen due to its reputation and data visualization capabilities and neither have been wrong.

Incentivized

The Cognos architecture is well suited for scalability. However, the architecture must be designed with scalability in mind from day one of the implementation. We recently upgraded from 10.1 to 10.2.1 and took the opportunity to revamp our architecture. It is now poised for future growth and scalability.

• Coming from an SQL based standard delivery model, Dundas BI has provided a fantastic rate of ROI.
• The preparation and delivery time has reduced from weeks to just a few hours.

Incentivized

• High ROI with well designed solutions. Supported by scalable deployment, robust security model and ability to create valuable content.
• High ROI where well designed data models can be deployed with a common metadata layer to a variety of users and use cases.
• High ROI in an environment that includes a variety of vendors and best of breed products within the overall platform

Incentivized