@JT: Is Great to have you to joining in the conversation, and your insight. I’m currently switching from MS Access to SQL Server 2008 (Express of Course) to improve scalability. I haven’t used Reporting Services but it definitely looks good on paper… Definitely the built-in integration with SAP Net weaver BI. Wish their report designer was more like Excel and less like a report writing tool like Crystal Reports (Darn!)… Nevertheless I’ll give it a try but your comment on the extra effort makes a lot of sense.
That said I’d love to continue the conversation. If you would please send me through e-mail the best way to contact you.

Cheers!

Jose

I am the co-worker that left Bulldog as well, that Tim speaks of… Tim hits the nail on the head. MS has an extremely promising tool and continues to make it’s way up the Gardner Magic Quadrant. With SQL Server 2005 and Office 2007, it is becoming a little easier to create solutions (I have done so at two separate companies now within the last year.) The reason why MS is a good solution is that most of us have MS in house or can purchase these tools cheaper and easier than other solutions because most companies would spring for MS tools a lot quicker than other systems. Plus, you have better support services for MS in house right now (am assuming). Furthermore, unless you have gone through a good period of prototyping of what you want to dashboard out (like Tim has done above), an expensive software system is of little use, especially if you haven’t exhausted MS capabilities.

Having said that, if you are ready to take the next step, MS can pose some problems.

MS is strong in pulling and pushing of the data (SQL Server 2005) but the delivery can become an issue. Dashboards in Excel can be super-robust by using some good technique, but to try and build on a global scale takes a LOT of time (imagine counting pixels all the time versus having WYSIWYG with another tool) and patience. In short, with MS as Tim stated, you are stitching pieces together in order to deliver a full solution and you will eventually need a lot of development to make things work. This is very similar to an open source solution. Therefore, even though the price is lower than a traditional BI tool (not by huge amount, either) on the outset, you have to weigh it with what resources you will need to bring in to make a full solution. Put on top of that MS notorious reputation for bugs, and I tend to lean over to a different solution (seriously, if anyone wants to argue this, have them spend every waking moment building a solution in MS and watch them sweat when implementing it!)

My advice, build prototypes in MS. You can change and update these pretty quick. At the same time, spec out the requirements you are going to need from the tool to get what you want. If MS can do this for you, by all means, use it, but if you find that it can’t, then time to start looking for another tool. The key is to have what you are trying to build, speced out already. If not, any good salesmen will show you a lot of things that have nothing to do with what you are trying to do!

I loved your post relating your journey. The year you spend finding the light switch is short in comparison to most of us. I use Excel for prototyping but also recognize its not for large audiences.

Somewhere in your blog you mentioned that in your new job you are looking for the platform to take it to the next level, and that were evaluating vendors.

I’d like to know your thoughts on MS reporting services (not Analysis… and the cubes). I’m playing with it and looking for a way to recycle the Excel prototypes into it.

Regards,

Jose

@Greg This was all done in Excel 2003, with no add-ons. The only macro in the spreadsheet is a “refresh all” button for refreshing the underlying data — it refreshes the data and updates one cell as to when the data was last updated (but is mainly there for usability reasons so that three different people in the company, who all have their ODBC connections set up properly, can refresh the data at any time). As I was working on this, a co-worker was busy developing a separate dashboard (external-facing) using Excel 2007 — we would regularly compare notes, and he would point out how much easier much of this was to do in 2007. The sparklines are just a small, very simple chart.

The balanced scorecard of Kaplan and Norton tracks business in the categories of financial, customer, internal processes, and learning and growth, where each category is to have objectives, measures, targets, and initiatives.

In most organizations, a significant change in strategies can occur in time for various reasons; e.g., perhaps there was a change in leadership. Hence, organizations that follow the balanced scorecard card and similar strategy-tracking strategies, as described above, do not have a structured framework for orchestrating and making analytically/innovatively derived decisions from an assessment of long-lasting value-chain metrics.

Balance is important; however, it can be detrimental to have forced balance, as described in the strategy-driven approach above. What is needed is a total integration and balance of selected metrics to avoid conflicting goals. For example, focusing on on-time delivery could lead to a reduction in product quality in order to meet ship dates.

Also, setting goals without a roadmap for making process improvement can lead to resource-draining firefighting activities; e.g., one week a metric could be red and the next week green just from process common-cause process variability – when nothing was done different to the process between weeks. Goals are not bad but goals need to be set after analytically/innovatively assessing the organization as a whole – to avoid organizational sub-optimizations. When a metric needs to be improved this metric is “pulling” for a process-improvement project creation.

As part of all effort, it is not only important to determine what to measure but also on how to report it, so that metric performance tracking leads to the most appropriate actions. This is achieved with satellite-level financial and 30,000-foot-level operational metric reporting. These metrics assess process predictability, and for processes that are predictable, a prediction statement is made.

A long-lasting value-chain measurement system can quantify how the organization is doing relative to its value chain. Current levels of value-chain satellite-level and 30,000-foot-level metrics can then be structurally analytically/innovatively assessed as a whole to build targeted strategies which lead to process improvement or design projects, for achievement of the strategies.

The Integrated Enterprise Excellence (IEE) system provides a methodology to avoid potential conflicts by basing the selection of all metrics on the enterprise value chain and assigning all metrics to the owner who is accountable for the metric’s performance. These metrics can be cascaded downward to lower organization functions, where they are assigned to owners who have performance accountability. With the IEE system basic value chain metrics will not change even if there is an organizational change. Only the ownership will change; i.e., the org chart is subordinate to the value chain.

@Andreas It’s almost like you have to be an apologist if you’re “just” using Excel at times. I’ve casually poked around on BonaVista Systems and XLCubed (and Crystal XCelsius for that matter) in the past, but never bitten the bullet and tried using them. Thanks for the reminder that they’re there!

>>As much of a fan as I am of our Excel/Access solution…it has its limitations (…) but I also view it as being the basis of requirements for an enterprise BI tool

With some tools like Micro Charts and the XLCubed you can turn your Excel Dashboard into and enterprise BI solution. XLCubed brings data connectivity to relational an OLAP sources, drill down capabilities and allows you to publish Excel Dashboard to the web as a scalable multi user web application:

http://www.bonavistasystems.com/Products_SparkLiner_Dashboards.html

http://www.xlcubed.com/en/Products_XLCUbed_Overview.html

Andreas

http://blog.xlcubed.com/