An Agile Approach to "Project Planning" Saves the Month of February

Bear with me for a minute, people. Let me say a few things about what happened to me last year, so you can better appreciate the pain I'm trying to avoid this year. In 2008:

• I created my own website, largely to promote my CMMI classes.
• The days I spent working on the site turned into weeks, which somehow morphed into a month, which ultimately became... well... I'm not sure because I didn't actually... uh... track my time.
  
• While adopting this heroic roll-up-your-sleeves, just-dive-in-and-do-it, you-don't-really-need-to-plan-it approach, I didn't sleep much, didn't exercise at all, got fat, and grew the world's ugliest facial hair.
• I canceled my first scheduled class (!!!) because I didn't finish my website in time to actually accept any registrations.
  
• When I finally did finish my site, it was crap. Just three weeks before class, only one student had enrolled.
  
• Harkening back to my mid-80's self, I unleashed a flurry of all-nighters. I massively re-worked the site, ignoring the small animals that were now nesting in my facial hair.
  
• I eventually did create a non-crappy site, which exists to this day (http://www.leadingedgeprocess.com). I attracted a respectable (but not profitable) eight students.
• As the year progressed, more and more people flocked to my classes! I shaved! I got skinny. I started dating super-models! (Okay, two out of three.... ) However, my site was virtually unmaintainable... and still is, as of today. I had boatloads of neato ideas for the site... but modifying it was a pain in the...
  

As it stands now, after months of finding more interesting things to do, I plan to attack my website problem. But I don't want to get fat again, and I certainly don't need to resurrect that small animal nesting place.

In my last blog post, I claimed I'd be applying the CMMI to marketing (my "project" of focus). Remembering that a project can be comprised of other projects, let's also treat the website as a project. (Consider it a critical subproject within my larger marketing project.)

Well, here's what I've done over the past few days on my web project:

• Decided to adopt an agile approach toward my web development, with a focus on timely delivery of useful, working iterations of the new site. In fact I now technically have a product backlog (that's agile-speak) of features I want in the site. I'll be conducting my first 30-day sprint starting next week, and at the end of that sprint I'll have a workable new site. In doing this, I'm largely addressing PP SP 1.4 Define project lifecycle phases on which to scope the planning effort.

• I've constructed a preliminary architecture to support my product backlog, and this architecture divides the site into nearly 40 distinct pages. This of course gives me a solid idea of the size of the product I'll be dealing with (as measured by pages). I've also associated a reuse multiplier with each page, because in some cases I can use portions of material that exists on my current website. So, because I have a handle on size and re-usability, I'm directly addressing PP SP 1.3 Establish and maintain estimates of work product and task attributes.

• I could go on here, but just trust me when I say I've addressed other practices such as PP SP 1.2 Establish a top-level work breakdown structure (WBS) to estimate the scope of the project (a no-brainer) and PP SP 1.5 Estimate the project's effort and cost for work products and tasks (easy, once I had size and reuse estimates).

So, how's this working out for me so far? Well, here's how a little agility has saved the day:

• My initial estimate for the new website, which will be hosted under a new (temporarily) top-secret domain name, was that it would take me 162 hours to build.
• I only have a month, tops, to build this site. After all, the clock is ticking on promoting my April CMMI classes!
• Problem: I simply don't have 162 hours of time available in the next month!
• Plus, I just received a query from a company that would like a private, three-day Intro to CMMI class this month. That would of course leave me with even less time, if I get (and accept) the gig.
  

So, what do I do? Well, not to get all agile-technical on you again, but I've simply removed some of the functionality from my upcoming Feb 9-Mar 10 sprint. These features will simply remain in my product backlog, to be addressed in a future sprint, if and when I get around to it. My estimate has now gone down from 162 hours to 119 -- which just may be doable, even with my other commitments.

Is there anything in this "agile" move that runs counter to the CMMI, either in letter or in spirit? Simply put, no.

A little planning, combined with a dose of agility, has helped to make my goal of a workable new website by March 10 much more realistic. Good news for me, because this translates directly into dollars.

What would likely have happened had I simply dove right in and started working on my website without regard to all this planning and estimating?

I think you know the answer. Just revisit the bulleted list near the top of this page. Goodbye super-models, hello small furry animals.