Another client went even further. A global investment firm with 200 offices decided to run Office 365 in its smaller offices, and then refused to continue to pay Microsoft for maintenance on the on-premises servers that were supporting those users. "They said, 'We're not going to pay for the perpetual licenses and pay maintenance on those products when we're moving the whole thing to the cloud.'" If you're going big on cloud, cutting back on on-premises support makes good business sense.
8. Don't negotiate a volume pricing agreement that accommodates the best- and worst-case changes in your seat count
You may have 50 users on that CRM platform today, but what happens if you grow -- or suffer substantial layoffs? How much does each incremental user add -- or subtract -- from what you'll pay?
Chances are, if you haven't negotiated that into the contract, you will end up paying far more than you should. You may not enjoy volume discounts as you grow, you might have to pay for empty seats and you could even end up paying more to add a new user if the provider's offerings change. For example, Salesforce.com discontinued its unlimited edition, which gave the user full access to all features and services. Now you may have to pay extra to give that user access to some features, such as Premier Support or mobile access.
In other cases the subscription pricing models of cloud vendors, most of which got started with small and mid-sized businesses, may not scale to the needs of very large enterprises. "When you expand from 100 users to 10,000 users globally, the [cloud] cost model just falls apart," Pearl says.
In those situations, the relative total cost of ownership implications between cloud and on-premises options may be clear only when analyzed over a long period of time -- say, seven years.
The more established players do tend to offer enterprise-scale tiers. But that brings up another issue.
"As you add licenses you should be able to add them to volume discount or user count tiers," says Wang. Rather than straight volume discounts, contracts can be constructed so that the business pays one rate for the first 99 users, for example, with a discounted rate for the second tier starting at the 100th user and so on.
The contract should be negotiated such that it will flex both ways, with per-seat charges changing accordingly as you peel away users from a given tier. "Get the tiering right because you need to be able to flex down as well as up," Wang says.
It's also important to negotiate on the ability to turn unused licenses into future credit, Wang says. "Typical approaches are to park licenses or return them for credit without impacting your discount levels."
Mergers and acquisitions create yet another set of issues. Consider the case of a professional services firm that started out with a 50-seat contract for its CRM service before acquiring another company and quadrupling the number of seats it needed. The business then had two contracts with two different rates, and the vendor demanded that the customer pay at the substantially higher rate after the merger.
The vendor wouldn't merge the contracts because the acquired business operated in another country, nor would it let the business renegotiate. "So they cancelled," Wang says, approached a new provider and negotiated terms that allow it to scale to as many as 1,000 users during the five-year contract.
Parting wisdom
Cloud-based services solve many problems, but contract complexity isn't one of them. As cloud services continue to proliferate, veteran negotiators say, businesses must build out or bring in the expertise required to avoid costly contract mistakes. Contract language is very different when you're leasing rather than owning the software, Wang says.
One great way to get up to speed: Attend workshops or bring in an expert to work alongside your own contract experts. Keep in mind that as the number of cloud contracts continues to rise, so too will the cost of failure. Cloud contracts represent an opportunity for a fresh start, says Wang. Perpetuating the mistakes enterprises have made with their on-premises enterprise software should not be an option.
This article, 8 sure-fire ways to screw up a cloud contract, was originally published at Computerworld.com.
Robert L. Mitchell is a national correspondent for Computerworld. Follow him on Twitter at twitter.com/rmitch, or email him at rmitchell@computerworld.com.
Read more about cloud computing in Computerworld's Cloud Computing Topic Center.
This story, "8 sure-fire ways to screw up a cloud contract" was originally published by Computerworld.