Yellowfin Review Summary
Yellowfin is something of a breath of fresh air in the over-hyped BI market. This is a platform that will address many needs in both large and small organizations, without the pretense that everyone with a display should spend all day playing with charts and dashboards. For those whose job it is to analyze data, Yellowfin provides a rich environment to do such work. For the rest of us who need access to dashboards and reports on an ad-hoc, as-needed basis, it will also provide the information needed.
So all the contemporary features we expect of a BI platform are supported by Yellowfin – charts, dashboards, mobile, maps, big data support, advanced analytics (via R), an in-memory columnar database for speed, multi-tenancy where needed, a semantic (or metadata) layer, strong collaboration features, connectivity to a wide range of data sources, and a web interface so that access is available on all devices with a browser. And all the dull things such as security, scalability and governance are catered for very adequately.
This is also a platform that might appeal to ISVs and those who might want to embed BI into other applications. The mechanisms exist to do these things with relative ease, and of course the multi-tenancy architecture (many distinct groups of users going through the same instance) will be attractive to those wishing to provide BI services in the cloud.
As mentioned in a previous review of Yellowfin, we consider this to be a no-nonsense, get the job done platform, equally capable as many other platforms, but with a more pragmatic approach – a good example being the use of the in-memory columnar database. Many suppliers tout this feature as the cure-all for high speed analytics. It isn’t, and for data that is rich in text or highly variable in content, an in-memory column database may well struggle. Yellowfin is keen to point these issues out, whereas many competitors do not.
This is a BI solution for organizations of all sizes that addresses the needs of a wide variety of users. IT will be happy with the governance features and the ability to set up a semantic layer. Analysts will find the visualization capability more than adequate, and casual users will find they can view dashboards and reports without a great deal of fuss.
Finally it has to be said that because Yellowfin addresses a broad set of needs it does not for example provide the extensive data visualization and exploration capabilities of products such as Tableau, Qlik and Spotfire. However for most business users there is no need to endlessly slide and dice data – most real needs are far more straightforward.
Yellowfin does all the data discovery and visualization things we have now come to expect, and does them in a way that will satisfy the vast majority of users. Dashboards, mobile BI, powerful mapping functionality, and collaboration with story-telling are all supported. A large number of chart types are available and the user interface is largely drag-and-drop and menu driven. The dashboards are interactive with filtering, drill down and through. The filter panel has pretty well unlimited capability, and again the interface is easy to use with various sliders and check-boxes where appropriate. Dashboards can be accessed simply by copying the url and embedding it into a web page.
The semantic layer and the ability to create federated queries across multiple data sources (including local spreadsheets) means that creating composite views of data is fairly straightforward. Finally Yellowfin supports some basic statistical functions and the ability to create calculations of varying complexity.
The storyboarding capability is worthy of mention, using pre-defined themes, so that formatting can be largely forgotten about. Mobile devices are particularly well supported by storyboards, and they support comments and bookmarking for collaboration. This is further enhanced by annotations and an inbox which serves to deliver alerts and any shared analysis. Workflows are also supported so that analysis can move through relevant quality controls.
Yellowfin supports a large number of data sources – database, files, applications and cloud services. There really isn’t much point listing them, and unless you are using some home brewed data format, Yellowfin will provide the means to connect. Spreadsheets particularly are treated as just another data source, and can be combined with any other data.
Once data has been included into the Yellowfin ecosystem it can, if appropriate, be loaded into the in-memory columnar database for very fast processing. As mentioned earlier, Yellowfin is quite realistic about what works and what doesn’t here. But for many types of transactional business data the columnar database will deliver very high levels of performance.
The meta-data layer allows an administrator (usually IT) to map data sources to a more business oriented representation. It is this that users access, without the complexity of obscure field names. Standard calculations can also be created here.
Governance and Management
There are numerous mechanisms within Yellowfin to ensure data is secure, that relevant permissions are well managed and that resources are allocated sensibly. Role, group and data level security is available and security mechanisms such as LDAP and active directory are supported.
All events within Yellowfin are logged, so that usage can be analyzed via bespoke or out-of-the-box audit reports. The architecture of Yellowfin is based on 64 bit Java, and supports high levels of scaling.
Embedding and Customization
Yellowfin addresses the OEM and ISV market through its ability to be rebranded and styled as necessary. In a multi-tenant environment each client can be styled separately. This is achieved through configuration options, cascading style sheets and image replacement. Security can also be integrated with any external authentication and role-based authorization mechanism – such as LDAP, Active Directory or custom security APIs – can be achieved with little effort. Yellowfin has a plug-in mechanism that allows users to code their own custom solution for authentication and course-grained authorization without changing the Yellowfin application or affecting Yellowfin support.
Integrating Yellowfin into third-party applications is easy with Webservices. Offered as SOAP services, Yellowfin Webservices enable integration with any programming language that supports the generation of functional stubs of WSDL. With Webservices, users can control and manipulate groups and users or run reports remotely. Yellowfin also allows users to embed either Jasper or BIRT xml report definitions that run within the UI.
There is certainly plenty of competition in this space. Yellowfin represents a good all-round BI solution, and is surprisingly open and extensible. Traditional BI solutions such as Microstrategy and Cognos are more capable, but less user friendly, more complex and more expensive. It does not have the advanced data visualization and discovery capability of Tableau, Qlik Sense or Spotfire, but is broader in its applicability and less focused on this type of functionality. In summary Yellowfin ticks all the boxes for general purpose enterprise BI.