Many customers use Gartner’s Magic Quadrant to get an overview of a particular software category such as Business Intelligence.  They typically use the “Leaders” quadrant as a short-list for selecting a BI vendor.

Gartner identifies leaders by counting the features their product set has, the number of customers they have, and the size of the company backing the software.  But these factors do not necessarily apply to you.

Instead, customers need to pay attention to how a product’s features match their requirements.  Its easy to have a career milestone of “bringing in IBM Cognos” or “leading the Oracle Hyperion implementation”.  Heck, the projects could die horrible deaths and it would still look good on your resume.  But that is not going to help your company succeed.

Did you know that only about 1/3 of the Business Intelligence products sold are from the 3 obvious leaders?  Two-thirds of companies did not buy the “best” products.

Leaders are great, but they may be more expensive and complex (and frankly more tempermental in some cases).  Targeted investments in best of breed tools from multiple vendors could actually be a less complex integration task than implementing one of the “big 3”.

If you are about to select a major vendor based on Gartner’s recommendation, think twice.  You may get better results for less money and time by broadening your search.

Learn how to make your dashboard project a success: DashboardSuccess.com

• many “versions of the truth” as people tweak numbers and send or save new versions of a particular Excel spreadsheet
• confusing spreadsheet formats – one senior executive we know of joined a new organization and asked for a simple consolidated Excel report from all divisions.  It took 20 people 8 weeks to produce it.  All previous Excel reports had been created in by each division in a different format and were completely inconsistent with each other.
• arguments about whether the data in the Excel reports are correct and/or current

So why do I recommend starting an Information Dashboard project using Excel?

The key to dashboard success is to implement as quickly as possible, show the dashboard to people and respond quickly to improve its contents and format.  An Excel dashboard prototype is the fastest way to do this, allowing you to quickly iterate through possible designs until you have a satisfactory result.  Then the engineering can begin.

The most difficult part of any information dashboard project is selecting the metrics that should appear on it.  This task is difficult because it requires managers to think deeply about the business they are in and what they are trying to achieve.  The environment most people are operating in is complex and fast moving – it is very difficult to extricate oneself from the noise and figure out what the core objectives are, and how to move forward towards them.

Metrics also must be chosen carefully to avoid unintended consequences – for example, causing people to suboptimize or game the system.

An Excel Dashboard prototype is a great way to run throught the metric selection exercise, but also to figure out which views of the data are most important.  For example:

• historical performance (trend chart or sparkline)
• comparison against target (traffic light, bullet graph etc.)
• comparison across metrics (usually not wise as the magnitude and direction can vary)
• comparison of metrics across locations and products (for example)

These are key aspects of successful dashboard design, and creating them in commercial dashboard tools may take more time than you have in your prototyping cycle.  Excel Dashboard prototypes are the best way to do this learning.

To learn more, please see http://DashboardSuccess.com

Interesting then, that Business Objects is positioning itself to be “the other horse”. Earlier I mentioned the rumors that BO is looking to be bought by SAP (here) . I also talked in my personal blog about how they are leading their sales process with Excelsius (here).

Now an interesting announcement of a partnership with Adobe. Specifically, Adobe and BO will:

• develop an Xcelsius Connector to Adobe LiveCycle Data Services (This allows Xcelsius users to get live streaming data from data sources using the Adobe back-end used by Flex developers. NOTE: instead of the BO back-end!)
• integrate Xcelsius with Flex (Flex is the programmer’s version of Flash – makes it much easier to build custom dashboards by using the Xcelsius widgets)
• enhance Xcelsius to export to the Adobe AIR platform (Adobe’s cross-platform “rich-client” desktop development environment)

This makes a huge amount of sense – Xcelsius whets the appetite of BI developers (and executives) for Flash front-ends. Flash/Flex developers get access to great widgets and the BO enterprise stack if they want it. Supporting the Adobe data platform is a surprise, but obviously an appeal to developers who are not ready to bury their heads in BO infrastructure.

Microsoft will be a gorilla in this space, but BO is rapidly positioning to become the other horse. Other BI vendors (read Cognos) beware!

Add to: | del.cio.us | digg

Information Dashboard Success

http://ezinearticles.com/?Reviewing-Your-Information-Dashboard&id=736949

Add to: | del.cio.us | digg

Information Dashboard Success

The author correctly identifies that:

“The biggest expense in setting up dashboards or scorecards is not in their creation but in their support. Acquiring reliable, quality data in a timely and error-resistant way is expensive. It means automating data capture instead of relying on manual chicken scratches and batch data entry.”

I particularly like the term “chicken scratches” – something all of us in the dashboard business are intimately familiar with. Although I continue to claim that hacks are an important part of the process of delivering dashboard iterations quickly, Larry is right that getting the back-end solid is always a bigger and more complex undertaking than we plan for.

“When considering setting up an operational dashboard or strategic scorecard, do not be dazzled by whiz-bang software demonstrations because those demonstrations use canned data, not your operational (or missing) data. And don’t get caught up in the hype and buzz currently surrounding dashboards. They are, indeed, powerful tools if they sit atop a solid database, but without the solid data to support it, a dashboard or scorecard is worse than useless… it is misleading.”

I love this straight-forward common sense approach – Larry is absolutely bang on with this observation. Dashboards are the ultimate “whiz-bang”, and it is easy to get distracted.

Add to: | del.cio.us | digg

Information Dashboard Success

It would be nice to see a cause and effect diagram showing how Lawyers * Utilization Rate * Blended Rate * Realization less compensation and other expense contributes to total margin.

Their dashboard commits many of the classic sins of dashboard design, but again, I very much like the metrics they have chosen. The day’s to cash process is extremely important and deserves prominence on the display. However, in order to be actionable, it would be useful to compare this performance over time, or by practice area / lawyer. One or more stacked column charts would be a better choice than the gauges.

Hours vs budget is also a very useful metric. Again, I want to be able to see which cases or lawyers are falling behind so I can have a discussion about why.

Some of this information is included in the Alerts view. I really like the idea of including unstructured, qualitative information on a dashboard in this form. However, these appear to be automated alerts – effectively the text version of a traffic light. I suspect there may quickly be too many of them to be useful and that it will be difficult to prioritize the ones that need attention. I also see no way to remove or mark the alerts that have been dealt with.

The 3-D pie charts (also a common dashboard sin) continue the analysis of time to cash displayed at the top. It would be better to tie these two areas together visually, and to use column or bar charts to show the number of accounts that are late. I would also be interested in how much money is sitting out there at 120 days, as opposed to the number of accounts.

Add to: | del.cio.us | digg

Information Dashboard Success

This news story explains how GE is now in the business of selling a “SmartCommand Dashboard” that displays household energy consumption.

Ok so its a glorified thermostat. But it is an indication of two things:

1. The power of the dashboard metaphor to describe how users of all types need to interact with complex data (not just executives)
2. The demand for interactive dashboards.

Information Dashboards to date have typically been one way portals – that is they display information, but you cannot touch it. True, XCelsius allows you to play with scenarios, but integration into back-end systems is not rigorous. Also true, executives much prefer to pick up the phone and yell when sales drop in a particular region, rather than “turn up the heat” on some digital display.

But this will change. Maybe not to address dropping sales, but certainly to adjust pricing, inventory levels or pull other “digital levers” to optimize the business.

Add to: | del.cio.us | digg

Information Dashboard Success

Your organization’s unique strategy to achieve long-term competitive advantage must be at the core of whatever scorecard or dashboard you implement. Whether you are confident enough in your ability to execute that strategy better than anyone else dictates whether or not you share it outside.

Financial

– Return on equity of 20-23%
– Diluted earnings per share growth of 7-12%
– Long-term shareholder value through increases in dividends and stock
price appreciation

Operational

– Productivity ratio of less than 58%
– Sound ratings
– Strong practices in corporate governance and compliance processes
– Sound capital ratios

Customer

– High levels of customer satisfaction and loyalty
– Deeper relationships with existing customers
– New customer acquisition

People

– High levels of employee satisfaction and engagement
– Enhance diversity of workforce
– Commitment to corporate social responsibility and strong community
involvement

Information Dashboard Success

Add to: | del.cio.us | digg

I’ve often said that a portal contains things of “interest” – sports scores, stock prices, weather in Cancun, blogs about your favorite car etc. A dashboard on the other hand contains information of strategic value to the execution of a manager’s responsibilities.

But what if the twain should meet? Henry Mintzberg wrote in 1975 (here) that executives are much more likely to consult external news reports, listen to hallway conversations, and rely on rumour or gossip than to act on rigorously prepared management information.

Managers are interested not in what has happened, but what is about to happen. They seek to understand the actions of their organization in the context of the environment in which they operate.

When orders for icecream suddenly drop in New Jersey, maybe its because there has been an unusual stretch of cool weather. Blockbuster sees a dramatic increase in rentals when the weekend forecast is rainy.

Maybe the weather does belong on the executive’s information dashboard. Maybe a dashboard should be a portal? The disadvantages of a portal for creating information dashboards are a topic for another day. But the fundamental idea that executives need to see internal company data in the context of their external environment is sound. Designing a user friendly dashboard with this breadth is an interesting challenge indeed.

Information Dashboard Success

Add to: | del.cio.us | digg

Note I say “attractively” not “effectively”. Stephen Few and Edward Tufte would have more than a few criticisms of the “non-data pixels” in this report and on this site.

But I think there is more to information dashboard design than optimizing the ratio of data pixels to non-data pixels. In an aircraft cockpit, the user has a serious incentive (not crashing) to use the data, so all that the dashboard must do is be effective at presenting information.

Information Dashboards on the other hand are only one alternative for executives to use when looking for business data. To encourage them to use the dashboard rather than call an analyst directly to ask for an Excel spreadsheet, a bit of eye-candy may be required.

The OnDemand site goes a little overboard in this respect, but I’ve often wondered whether carousels would be useful on an Information Dashboard and here is an example that is worth looking at.

The other thread here is the notion of integrating external data into your dashboard so that internal data is shown in context. Now there is a topic that merits some exploration.

Information Dashboard Success