Life in the Jungle

It's hot, it's muddy, there's all kinds of buzzing noises and itches that need scratching, and I'm sweating in places I didn't know I had.  No - supporting a "Dashboard" in MS Access is not the kind of vacation they advertised.  Yes, this is an update on my Guerilla BI strategy.
I've barely gotten any time to work on my Data Warehouse in the last six months - maybe 5% of my time.  But there has been some progress.  The last time I'd mentioned by Guerilla BI tactics, I had my nightly ETL process 95% completed - with respect to extracting the scope of data that I'd intended for a "first step".  As with all warehouses and BI solutions, and any software solution for that matter, the scope grows organically as you complete phases in your project.  Since then, I've tweaked the ETL a little, and fleshed out the SSAS cube.  Some of my calculated measures were very tricky to implement - and I'm not entirely sure why.  Now that I think of it (hindsight) I think I'll post a comment over at the SSAS forums for the experts in AS/MDX to look over.  They may have some suggestions to improve my implementation, or some questions about my sanity for constructing my members the way I did.  Regardless, I have my cube set with a good set of dimensions, "regular" measures (like COGS, margin, units sold), and calculated measures (like Turns and GMROI).  In order to get the Turns and GMROI, I had to add an inventory snapshot fact table to my warehouse and use some conformed dimensions.  I still have to work on the logging aspect of my ETL, but I haven't gotten around to that yet because I don't yet have a server provisioned for me.  Yes, I'm running my warehouse entirely on my own dev machine.  But then, I'm happy with that at this point, because there's no significant pressure from the potential user base yet - and it's a work in progress.
Now that we're talking about the user base... There are about ten people in the company who are aware of my "subversive" activity... and yes, one of them is my boss.  He's been encouraging me to spend that 5% of my time, knowing full well that it will (eventually) return the effort in spades.  Every time he gets a complex reporting request from the other executives, he tries to pass it my way to fill.  In part because he should, he's really too busy to be extracting data manually and constructing reports (although he's eminently qualified to do so) - and in part to "test" my cube.  Those tests have been very beneficial - they've required some tweaking and extending of the architecture, all well in advance of it really being viewed by "users".  On the last such inquiry for information, I was able to provide them information in minutes that they simply could not get from anywhere else in less than 8 person-hours.  We let the two execs requesting that information know where it was sourced from.
So I'm partially out in the open - the foliage in the jungle is thinning.  Whether that will mean true BI will emerge into the flowered meadow to take over from the bloated and diseased Access system, or just be exposed to some cost-cutting carpet-bombing remains to be seen.
On the technical side, I still need to get my server provisioned (it's in the pipeline), set up my deployment architecture (I have a rough plan), and add budget information to my warehouse.  On the "people" side, I need to get one or two mid-high level managers involved to understand and critique the Excel pivot-table experience of using the cube - and that will hopefully start the viral "win their hearts and minds" war.