Logistics and Supply Chain Dashboard

Blue Sky Logistics offers a dashboard that gives insight into the logistics and suppy chain function.



The prominence of "Alerts" here is interesting - as I have observed elsewhere however, alerts should be qualitative information entered by humans. If machine generated, they can typically be represented in a different form - perhaps with traffic lights or other indicators of urgency.

Cash-to-cash is an important thing to measure regionally - I wonder how they would manage a very deep hierarchy of regions.

For order fullfillment cycle time, I would tend to use a bullet chart if horizontal display is desired. To my eye, its difficult to translate the bars into their meaning - another possibility would be to display them as columns. I find the Return on Supply Chain fixed assets column chart easier to read than the order fullfillment bar chart.

Hopefully, there are options to display periods with better labels than "1,2,3,4,5" on the perfect order fullfillment chart. I also want to see this broken out by facility etc. Sparklines?

Add to: | del.cio.us | digg

Information Dashboard Success

Real-estate CRM Dashboard

I came across Bluetree Direct today. They show a nice little CRM system for Real-estate agents. The dashboard portion of the product appears to have been created using xCelsius.I find it interesting how XCelsius dashboards show so much better in an interactive demo than in a screen grab. The engagement really seems to be about the animation of the charts - as a report writer, XCelsius is only average.

There is actually not much to like in this dashboard. Clicking on the "Source" pie chart likely changes the data shown in the other three charts, but this will cause you to have to remember the Email results in order to compare with the Direct Mail results which is what you want to compare. It would be much better to offer a view in which deal size, sales status and other information can be compared across the marketing "channels".

I do like Bluetree's full range service, including graphical template selection, automatic printing and direct mail along with email and other forms of marketing at the click of a mouse. The dashboard could use a bit of work.

Add to: | del.cio.us | digg

Information Dashboard Success

Dashboard: Rear-view mirror or Heads Up Display?

Chris Carren who hails from Business Objects and is now general manager of Microsoft's Office Business Applications division (interesting in its own right) released an interview yesterday concerning Microsoft PerformancePoint. He says:

There's also been a huge evolution in how customers want to think about business intelligence. They are showing an eagerness to move away from the traditional methods of sharing information, like sending reports out via e-mail that capture what transpired the previous quarter or reporting period. Instead, they want a metric-centric view via a Web-based scorecard or dashboard that focuses more on what's happening now, and even more importantly, what likely will happen next week or next month.

That last sentence is interesting. Although predictive analytics firms have been around for years, they have achieved little penetration into the mainstream market. Witness Cognos's acquistion of a product "Forethought" about 7 or 8 years ago that was promptly buried.

Today, I think Chris is right on the money due to a few key shifts in the market.

1. The demand for forecasting and enterprise planning solutions has picked up in recent years, as people try to move off spreadsheets and into something more integrated, accurate and controllable. Sarbanes-Oxley contributes to this among other things.
2. Increased focus on data-based decision making. While many organizations do still struggle to get accurate data, a substantial number now have this under control and business intelligence is part of their management process.
3. The "Competing on Analytics" trend is just beginning to take off. Companies are starting to realize the power of statistical analysis, experimental design and forecasting to improve their business results.

As the "quants" get involved, SAS and SPSS get a big push. But amateur quants also need tools - hence Excel 2007 SP2 in which users can run the data mining features of SQL Server Analysis Services 2005 from their desktops.

I first heard the phrase "driving by looking in the rear-view mirror" in one of the early Balanced Scorecard books. Kaplan and Norton used this as an analogy for managing your business by looking at past financial results only. They encouraged organizations to seek "leading indicators" that would drive future financial results.

Still, scorecards and dashboards typically only display the past, even if they include leading indicators. What Chris is talking about (and Thomas Davenport for that matter), is using the past to statistically predict the future, and display that on the dashboard.

I'm not sure dashboard is quite the right image for this trend. Instead, I think about the heads-up display. Managers are razor focused on the airspace ahead, with analytical and predictive information thrown up on the windscreen. A warning light indicates something bad might happen. A predictive chart validates whether the executive intuition matches historical projections. All in real-time with reaction times down to the hair trigger.

Compare that picture to the stodgy data warehouse and BI departments and vendors that litter our corporate battle field...


Add to: | del.cio.us | digg

Information Dashboard Success

XCelsius: Eye Candy for BO

I had the opportunity to sit through a Business Objects sales presentation earlier today. Here are a few things that surprised me:

First, I had missed the significance of their acquisitions in the data quality and ETL space. They now claim to be competitive with Ascential and Informatica for ETL, and to have acquired the number 2 player in data cleansing. I have no idea if this actually works on the ground - whether the products integrate, whether they are as good as they say, or whether BO actually knows anything about these areas. However, it does make for a compelling sales pitch.

Text analytics is an area I had heard they were making a foray into. The sales rep pushed that hard and talked about the capability to generate automatic reports containing structured and unstructured information. Again, no show - just tell, so I don't know how much is smoke.

XCelsius is a sexy product, though I have numerous concerns about it as an effective dashboard or analytical tool. I should not have been surprised that BO sales is leading 100% of their demos with it. They surf over the data architecture issues, and spend all their time showing the "interactive" Flash dials, charts etc. I couldn't believe they showed almost nothing else in their suite in a 1 hour demo!

Desktop widgets are cool. Essentially they can put any BO content into a desktop widget, which can then occupy portions of the user's desktop with updates in real-time. I had considered doing this with Google Desktop, but don't think they have much corporate penetration. BO's offering is again the kind of eye-candy that account execs can use to overturn any kind of real objections and get buyers emotionally involved.

Information Dashboard Success

Add to: | del.cio.us | digg

Information Dashboards for Collaboration

Information Dashboards are typically thought of as individual productivity tools. Stephen Few's popular definition of a dashboard implies this:

A dashboard is a visual display of the most important
information needed to achieve one or more objectives;
consolidated and arranged on a single screen so the
information can be monitored at a glance.

(from Dashboard Confusion Revisited)

But the real value of a dashboard is the change in the nature and quality of collaborative interactions that occur betwen managers and executives as a result.

Many executive meetings are spent in useless poltical squabbles over who has the correct data. An information dashboard which all agree contains the "single definition of the truth" ends this unproductive conversation, and allows a new and more useful conversation to develop.

Initially, the broad availability of data (even for detailed performance of other organizational units) can cause some fear and push-back. Yet when all agree to "see the world the way it is and not as they wish it would be", they become able to contribute their creativity as a team to improving the whole organization, and not just their individual silos.

The upshot of open collaboration around a trusted single source of data is that organizations can learn faster about what works and what doesn't work. Each action in a remote organizational unit can be considered an experiment - if it works, it can quickly be deployed to the rest of the organization instead of being blocked by parochial leaders looking to advance their own careers. If a failure occurs in one part of the organization, the diagnosed cause can ensure that it doesn't occur anywhere else.

Information Dashboard Success

Add to: | del.cio.us | digg

2 x 2 Charts for Vendors, Pipeline and Time Management

A 2x2 chart is an xy graph which is divided into 4 quadrants. Usually data points in the upper right quadrant are designated "good", while the others are some variant of "bad". Here's an example: The Gartner Magic Quadrant (from MediaProducts.gartner.com)
This is actually a useful view for certain Information Dashboard requirements. For example - it can be a useful way to look at the sales pipeline. Instead of just looking at probability of close, or dollar value, you can see both for each prospect. Those in the upper right deserve more focus than those in the lower left.

Stephen Covey used one of these charts many years ago in his time management system (see it here), arraying urgent tasks against important tasks. In this case, the idea was to make time for important non-urgent tasks. What a fabulous idea for all of us to be frequently reminded of...

Information Dashboard Success

Add to: | del.cio.us | digg

7 Sins of Dashboarding

Here's the list - don't ever do them:

1. Use gauges
2. Use pie charts
3. Use trend arrows
4. Sort metrics based on status (e.g. traffic light color)
5. Use radar charts
6. Use bubble graphs (where the size of the bubble matters)
7. Increase data pixel to non-data pixel ratio without a clear understanding of the "story" being expressed by the dashboard

More on this in the book: Information Dashboard Success

BI Market grows by 11.5%

IDC reports that the BI market grew to $6.25 billion in 2006, with ongoing consolidation among vendors.

http://www.sas.com/news/analysts/idc_bi_0607.pdf

Interestingly, IDC concludes that the marketing messages used by BI vendors are failing to increase the size of the market.

IDC says that interest in advanced analytics (see Competing on Analytics) is growing, despite the fact that analytics as a share of BI revenue has actually dropped from 20.5 to 19.9 over the past 3 years.

Floyd
Author: Information Dashboard Success