The Failure of IT Projects

A found this article on Why So Many Big IT Investments Do So Little for Shareholder Value via a great blog: Drakeview and was once again floored by:

He cited figures from Gartner Research showing that “on average, 20% of the corporate IT budget is spent on initiatives that don't achieve their objectives. That means $500 billion of bad investments.” In the U.S. alone, he said, total spending on Customer Relationship Management has reached $10 billion, “but analysts estimate that more than half of CRM projects fail.” In addition, more than 90% of companies are dissatisfied with the results of their Enterprise Resource Planning implementation.

Being a small software company, this simply stuns me. I can't imagine wasting 20% of my budget to implement a piece of software or a suite of software that didn't either a) provide returns to the bottom line or b) make my team's tasks easier. The author believes that the reason for this boils down to 2 points:

“One, it's difficult to sustain a technology-based competitive advantage; two, companies often lack the management discipline they need when evaluating technology proposals.”

The first point makes perfect sense for two reasons. First, a technology company, whether it's CaseySoftware or Oracle wants to make a good return on its own investment. Therefore, there is a fundamental drive to take the code, product, or service developed and honed for your organization and modify it for the next organization and charge much less. This is why most CaseySoftware projects – like the soon to be released Microsoft Project to dotProject importer – have built in agreements not to recommercialize or publicly release the specific code for at least three to six months. This gives our customers the opportunity to completely integrate it into their operations and make further improvements.

Second, if a particular organization jumps ahead with a huge advantage, its competitors *must* work to close the gap as soon as possible. Look at something relatively benign like “Member Cards” at grocery stores. A few years ago, only a few places used them. After it was seen how much marketing and purchasing information could be gleaned from them, every store uses them. For example, we recently completed a web-based Lead Management System for a local group of “house flippers” (people who buy homes and attempt to sell them quickly at a large profit) that tracks the property through the entire process and calculates the costs and profit in realtime. We have already been contacted by three similar groups…

The second point: “Companies often lack the management discipline they need when evaluating technology proposals” strikes me as humorous. As mentioned in Carnival of the Capitalists yesterday, a blog cannot solve all your marketing woes. I've gotten numerous proposals describing how a business plan with technology X is “going to go crazy.” If someone told you they were going to become a master carpenter by buying a pneumatic nailgun, you'd think they're crazy. Why do we accept this with technology?

Finally, most organizations don't know how to measure the results of their implementation. How much time has the new system saved? Are manual processes improved/eliminated? Do customer invoices get out sooner? Do vendors get paid faster? What are the old and new error rates of invoices/payments? Are sales increasing? Are infrastructure costs lower?

In my last non-CaseySoftware firm, I worked on the development and rollout of a huge Ticket-based Time Tracking System which handled Time, Materials, and Equipment tracking and billing. Within 60 days of deployment, we found three interesting measures.

First, invoices were being sent to customers within 7 days instead of 45. This resulted in payment often being received in the same quarter after it was billed as opposed to 75 days after the work was performed.

Second, payroll took two people one day each instead of one person fulltime and was therefore no longer on the Critical Path for payroll processing. [That one person was re-assigned to other tasks that had been going undone.]

Finally, Material purchasing was able to be coordinated across crew and customers to negotiate better prices with Vendors.

Without having these sorts of metrics, success/failure can't be measured and therefore an investment cannot be clearly evaluated.