This week, SAP announced LAVA “Lightweight Applied Visual Analytics as their new approach for data visualization within the SAP Business Intelligence portfolio.
According to SAP’s Anita Gibbings , “This new and exciting Design Language provides a basic set of guidelines, forms, and techniques to be used across products to ensure best practices, consistency, and reliability”
I believe LAVA is a step in the right direction for consistency to make visual analytics from SAP more approachable, and mobile friendly. There are some really nice elements introduced in LAVA that we have not seen across the SAP BusinessObjects portfolio before; specifically Lattices.
What LAVA Is
Here are my takeaways from Anita’s Article on SCN:
LAVA is a more visually appealing design-wise for presenting and organizing quantitative data.
LAVA presumably could provide a consistent visualization specification across all analytics tools.
LAVA provides modern, device appropriate paradigms for interacting with information
LAVA’s provides various visualization containers and new charting mediums for organizing more information onto the screen
LAVA will force product groups in SAP to remove some of the visual interference often attributed to some of SAP’s more Flashy visualization libraries.. Perhaps exploded pie charts will finally go away for good?
What LAVA Isn’t.
LAVA is an incredible step forward for SAP’s visualization approach but not necessarily an evolutionary leap forward in visual analytics. My belief is that a bar chart is a bar chart… Whether large, small, micro, drillable, filterable, etc. No matter how you orient a bar chart on a page or how fast you can swipe across pages of charts, it will not necessarily enhance your ability to assimilate information. It is the engine behind the bar chart and the right application of the visualization itself is where you derive value. With that said, the verdict is still out until SAP puts LAVA enabled analytics into the hands of business users.
LAVA is not a new product or framework or API. It is a set of design principles that presumably will be integrated across multiple BI tools.
In short, LAVA by itself is not going to revolutionize decision making. The key for revolutionizing the actual decision making process is predictive analytics. For example, if Predictive Analytics is the engine, CVOM is the visualization framework (technology), and then LAVA is the design specification for which the visuals are implemented.
Lets get Picky About Visualization
Here is an example illustrating how LAVA has moved from a flashy, gradient rich visual, to a more sobering view. As I note below, you have to consider that the visualization skin on the right has not been updated by SAP since iPhone hit the shelves, which is indicative to where design was 6 years ago.
My main problem with this LAVA example is that it lacks any visual indication of what is good or bad. If that is not the intended use or that feature was disabled, I can rescind this critique.
Onto pure visual design, I would argue that the LAVA sample on the left presents equivalent visual interference with redundant grid lines and blue lines extending from the tip of the bar chart. Furthermore, the stylized font itself while narrow to provide more real estate is actually difficult to read. Finally, if the bar colors for Expense type is not consistent across the whole page, there is no need for each bar to be a different color. I am pulling one sample out of context as typical data visualization critiques do.. Lets wait to see what happens when the data visualization gurus get ahold of LAVA…
What I hope to see LAVA do..
My biggest gripe about SAP’s newer analytics tools is that they lack a visual medium for indicating what is good or bad. Visualizing “Performance” is something that customers want, but SAP BusinessObjects’ specialty is visualizing data. LAVA seems to be a great step in the right direction as long as the visual design language provides mechanisms for highlighting: outliers, tolerances, bands, moving averages, etc. All of these fancy words equate to business people identifying what is actually important.
For example, what if expenses have shot up substantially month over month? Does LAVA visualization account for business users having the desire to flag visually what expense had the greatest change month over month? If not, SAP is still stuck in the mindset of presenting data rather than presenting performance. What we need is LAVA along with incremental steps forward in the applications to make visualizing performance more effective. This is why the defacto BI tool is and still will be dashboards.
“However, to tell a story you need context (Text). If we are designing the “next generation” of visual storytelling, we must identify what is good, what is bad, and provide context to what it all means. You can call it a report, dashboard, or storyboard, but an assembly of charts on a page describes data, not necessarily meaning. That is the next evolutionary leap that has to take place here.
Predictive analytics is the end game for most organizations at this juncture but there is a huge gap between descriptive and predictive.. Unfortunately LAVA while cool looking just makes the the descriptive piece look better, not communicate better.”
My assessment is based purely on the article examples and the buzz I had heard previously about LAVA. However, it is also based on countless conversations with BI practitioners who have succeeded and failed implementing BI. Business users they do not ask for a prettier picture to describe their data… They ask for solutions that will allow them to accomplish tasks faster, reduce mental juggling, and reduce hunting for information. At the end of the day, I am thrilled to see how SAP pairs the new innovations for analytics with this new specification for organizing data on the screen, but the community will need to test first hand to see if the LAVA magic lives up to the hype!