Maintenance: 7% or 22% — or something in between?

InformationWeek last week published a very interesting article on the options available to software vendors as they determine how to price maintenance. In case you missed it: SAP customers revolted earlier this year, refusing to support a maintenance price increase for new customers until SAP proved the value of elements of the maintenance contract. SAP reached agreement with its user group to study the problem and work out a solution, while rates remain unchanged. Vendors, of course, rely on maintenance fees for much of their profit, so Wall Street hates any thought of maintenance fees going down. Maintenance pricing is also a competitive issue, as 10 years of maintenance at 20%/year mean that a company essentially pays for its installation three times before considering a significant upgrade or vendor replacement.

Since SAP’s closest competitor is Oracle, the InfoWeek article postulates 3 scenarios that incorporate the financial, investor and competitive aspects of the problem: no change (both vendors continue to charge 22% per year); one of the competitors offers a tiered maintenance structure (7% – 22%, depending on level of service), drawing the ire of Wall Street but possibly winning new customers; a third party enters and radically changes the game.

Clearly, scenario #1 is unsustainable in this economy. Companies are letting maintenance lapse rather than paying for something they feel delivers no immediate value, even if the employee using that license is still employed. This is tricky for the customer (how much will it cost to turn the license on again someday?) and for the vendor (who’s going to cancel today? how do I manage costs within an unknown revenue stream?). Too, a lack of “give” by the vendor can make customers feel abused and could lead them to explore alternate vendors in the future.

InfoWeek’s scenario #3 is based on the fact the ERP software is often implemented by a third party consultant and that these consultants could eventually replace SAP and Oracle in offering tech support. I don’t see this as all that likely in technical computing, although it does happen with enterprise PLM user.

I’ve always favored scenario #2: offering a tiered system of maintenance that allows buyers to select their level of service and pay for what they use. Requesting service above the current contract level would allow the vendor to charge a premium but customers won’t feel abused if all they ever want are the bug fixes. Of course, this kind of change would require vendors to rethink the value offered by their maintenance programs and, in essence, prove that value to the customer base. What would it say if very few went for the top-priced tier of support? Vendors would also have to adapt their business models to a far lower level of recurring revenue, offering less of a cushion for risky new product development. New license revenue would have to make up for any deficit, leading to the real question: how many new buyers would be lured by this approach?

What do you think? If you’re a vendor, 22% is perfect. But if you’re a buyer, would 7% sway you to consider an alternate to your current supplier? Even if that 7% ups their business risk? Email me at monica AT schnitgercorp DOT com.