Warning:

  • We are in a deep economic slump.
  • The economic future is more uncertain than ever.
  • Predictive, plan-driven waterfall processes for IT Program and Project Management can be dangerous to your ROI.

The first two warnings are obvious, the third is less so. But, as this article explains, selecting and executing large IT software projects using plan-driven, waterfall processes limits flexibility, and defers the recognition of benefits, both of which are dangerous to Corporate ROI in this weak, unpredictable economic environment.

Part two of this series will cover how agile processes provide the options to abandon, switch, defer or grow a project, providing the responsive, flexible approach that IT needs in today’s difficult and uncertain times.

Poor Economy and Increased Uncertainty

Much of the world is the grip of a deep economic slump, and future conditions are uncertain. Economists in the U.S. had first been concerned about inflation, then with a deflationary spiral, and now hyper inflation may be on the horizon—due to Federal Reserve pumping in liquidity.

As an example of uncertainty, on July 4th 2008 Americans were celebrating the Independence Day holiday, but they were not cheering the price of gas, as crude oil was trading at over $137 a barrel, and many were predicting that a $200 a barrel oil price was just around the corner. Seven months later, in January of 2009, crude is now trading in the forties– the price fell off a cliff, with a decline of over 70% in six months (U.S. Energy Information Administration chart).


Other commodities have similarly dropped precipitously leading to concerns about a deflationary spiral (e.g. your house dropping in value 75%). In response, central banks are flooding the financial system with money, and this liquidity may result in hyper-inflation (e.g. your food costs rising 25% in one year) after the economy rebounds.

Making decisions on large IT Projects is difficult when you don’t know if:

  • Your competitors will be cutting prices or going out of business?
  • Which of your customers will still be able to purchase from you?
  • If the cost of raw materials will spike up or down?
  • If you will have access to the funds needed for large IT investments?

Moved to Survival Mode, Still Using Predictive Processes???

As a result of this uncertainty and the poor economy, many companies have moved into survival mode, scaling back plans, cutting projects, calling in lines of credit, reducing headcount, etc. But, many still rely on predictive, plan driven processes for selecting and executing projects.

These processes assume it is possible to accurately forecast a benefit that will be achieved in the distant future and as a result they don’t provide the flexibility or responsiveness required with the increasing levels of uncertainty most business’s are experiencing today (granted predictive processes can make sense for some classes of projects in more stable environments).

Let’s review a typical plan driven process for turning an idea into an approved project that gets delivered. We will call our hypothetical project, Project-X. The idea for Project-X was hatched in May and then discussed informally using PowerPoint slides. Then a set of formal documents describing and justifying the project were created, including a ROI (Return on Investment) analysis.

A basic ROI formula is: ROI = (Benefit – Cost) / Cost

The ROI showed the ratio of forecasted gain relative to the forecasted investment. Project-X also calculated a net present value of the ROI, to account for the time value of money, and documented non financial benefits in respect to customers, staff and internal process, to provide a comprehensive basis for project approval.

In August, Project-X received departmental approval, and it then became the centerpiece of a Roadmap that was presented to a Corporate Council in October. With some additional work in November and December, corporate approval was gained in early January of the following year.

At that point, Project-X had been bouncing around for over six months. The project began executing in February using rigid: plans, requirements and designs; in a sequential waterfall process.

Unlike most projects, Project-X was executed to perfection, delivering exactly as planned, with no budget overruns, delivering all requirements exactly as they were originally requested.

On Time, On Scope and On Budget, but a Financial Disaster

Project-X was delivered on time, on scope, and on budget. From the Project Manager’s standpoint, it was an incredible success. But, unfortunately for the company, Project-X was a financial disaster.

  • The requirements were completed just as they were originally asked for, but they made little sense at delivery time as the business environment had considerably changed.
  • Competitors had gotten in ahead of Project-X, stealing away most of the forecasted benefits.
  • During the 18 months that Project-X was being executed, capital became increasingly scarce and costly. The company had so much money tied up in Project-X that they couldn’t work on other key initiatives.

The scenario is hypothetical but the outcome is typical.

Questions:

  • Is it appropriate to base project approval on forecasted benefits that won’t be achieved for another 18 months?
  • Is it appropriate to use a process that requires two years to convert an idea into a production release?
  • Is it appropriate to use a process that provides little in the way of flexibility (i.e. options) for course correction in response to changes in the environment?

There are classes of projects, such as some in the DoD, where you can answer “yes” to the questions, but, for most corporate projects the obvious answer to these question is “no”. Despite this, many organizations continue with predictive plan-driven approaches whose lack of flexibility and slow speed to market often result in failure.

Development Processes Must Offer Options

In the book Extreme Programming Explained, Kent Beck describes one way to view the economics of software project management, as a series of four options:

  • Abandon – You can cancel a project, but still gain some value.
  • Switch – You can change direction, the more often that requirements can change the better.
  • Defer – You can wait until the situation is clarified before investing, the longer you can wait the better
  • Grow – You can grow quickly if the market takes off.

Kent goes on to explain that using a project management process that supports these options becomes increasingly important as the level of uncertainty increases. I agree with Kent’s statement and believe that now, more than ever; agile processes are the appropriate choice for many corporate IT Software Projects.

Summary: Plan Driven Approaches Aren’t Flexible Enough

With the increased uncertainty inherent in today’s business environment, predictive plan driven processes for selecting and executing large IT Software Projects will often fail to deliver the forecasted business value. These approaches focus on meeting the scope, budget and schedule constraints that are defined at the start of a project, but they delay the time to the recognition of benefit and they don’t adequately account for a changing business environment.

Part two of this article will further define how iterative, agile processes, deliver the flexibility (i.e. support options to abandon, switch, defer or grow), and the speed to market, required in today’s economy.

Questions?