I am sure many of you have been through the same experience as mine: We put together a thorough due diligence checklist when we selected the software. We formed a cross-functional selection committee, started with several candidate software companies, winnowed down the choices, performed an in-depth
Back before I knew any better, I followed the vendor's lead and included in the contract all the licenses I thought I would possibly use for the rest of my natural life, as well as all the specific products I thought we would find a reason to use. (Sure, my warehouse manager is very happy with the incumbent warehouse management system, but once he sees what a killer deal I got on the WMS module from the new vendor, I am sure he will want to use it.) Besides, the vendor is telling me I will get this great discount only if I sign before the end of the vendor's current financial quarter. If I wait until the new quarter starts, the vendor will have less motivation to give me such a great deal.
After taking this approach a couple of times, I realized what terrible mistakes I was making. First, things changed so much that I never used any of those additional, I-will-need-these-at-some-point licenses. Second, I really didn't need to include the agreement licenses for modules that were not in my near-term planning horizon. Third, I did not need to take advantage of quarter-end pricing to get the desired discounts. Once, I missed my deadline for an end-of-the-financial-quarter deadline. Guess what? At the beginning of the next quarter, when I was finally ready to sign, the discounts were identical.
In short, I was wasting an incredible amount of money to get better pricing for licenses I was never going to use. Since that time, I have taken a completely different approach to software licensing. Let me share some specifics of that approach:
Buy only what you know you will need. The most recent research I can find indicates that about 40% of the licenses we purchase end up as shelfware, and are never deployed. It would take incredibly deep discounts to overcome the pricing disadvantage of buying 40% more licenses than we need.
When negotiating the contract, agree to pay maintenance and support only on the licenses that are deployed. And, to take matters even more firmly in hand, the annual maintenance clock starts when the licenses are deployed initially (not when you sign the contract). Thus, if it takes six months to implement the software, your annual maintenance payment is due twelve months after you go live.
To deal with uncertainty about how many licenses you will need, negotiate for a discount protection clause -- ideally one that never expires, but the best one I ever received was for 18 months -- that lets you acquire additional licenses at the same discounts you negotiated for the licenses you purchased in your contract.
Make rational decisions about maintenance and support. These contracts cause us, in effect, to repurchase the software every four to five years (or in the case of one very large, well-known-for-licensing-complexity vendor, every three years). We pay for this even if the software rarely changes or the support is substandard. In some cases, this is insurance I simply don't need. For some of our applications, we have decided that we no longer need support -- the applications are stable, and we have internal expertise. For these, we have opted out of our support contracts. In other cases, we have decided that the application is close enough to end-of-life that we are no longer considering upgrades, and so have opted out of maintenance contracts.
More contract tips
In the case of the vendor with the three-year support and maintenance agreement, we looked at its product release cycle (an average of every three years), new features and functions (not any that we are dying to have) and our adoption of their new products (longer than their three-year release cycle); and we have decided we don't need to repurchase their products every three years. We save money by purchasing the software fresh, when we decide to move to the next version -- if ever. For us, these are rational support and maintenance decisions.
Some experts claim that we can do little to improve our IQ. I figure that when it comes to software licensing, I might not have a higher IQ than I had some years ago, but I sure am a lot smarter.
Niel Nickolaisen is CIO at Western Governors University in Salt Lake City. He is a frequent speaker, presenter and writer on IT's dual role enabling strategy and delivering operational excellence. Write to him at firstname.lastname@example.org.
This was first published in July 2010