|
Contact Us! |
A service level agreement (SLA) represents a series of commitments by a vendor to a customer. SLAs do not have to be complex or mysterious. However, many Application Service Providers (ASPs) use arcane language and incredibly detailed standards to make the SLA appear to be a serious, life or death obligation.
Short answer: they aren’t at all.
ASPs make SLA commitments to convince skeptical customers to hire their services. If customers automatically trusted vendors--i.e. if vendor performance could be more predictable--SLAs would not need to exist. In a high tech industry recovering from highly publicized software project collapses and equally challenging eBusiness euphoria, any entity seeking the services of an ASP must negotiate an SLA.
This particular series of articles seeks to encourage best practices, standardize usage of SLAs in the ASP industry and shed light on some of the more shaded SLA canyons. It is our hope that these pieces will help every prospective ASP customer ask valuable questions of the ASPs they evaluate.
Any SLA must provide answers to five basic questions. Please note that in our experience, most of these agreements address the first three points and grow very quiet on the final two:
Service contracts for construction, facilities management and utilities always contained service level provisions. These 'Classic SLAs' contain requirements, such as:
These concepts are familiar to high technology negotiators who have a working familiarity with SLAs. Mistakenly, however, many high tech professionals believe that the SLA originated in the high tech industry and that it operates under an entirely new set of rules.
They are half right. SLA rules have changed for high tech, but mainly because people on both sides of the negotiating table do not fully verse themselves on what constitutes an acceptable level of service and how to capture those standards on paper.
Many ASPs originated in the information technology and telecommunications industries. Their perception of an SLA comes from the telecommunications industry, which made the SLA--and ways to escape their provisions--into a fine art as early as the 1970s. Over time, terms and conditions unique to the telecommunications environment migrated into the mainstream of technical contracting. Proponents of the ASP model who sneer at misuse of technical terms, commonly misuse legal, contracting and general business terms. Misuse becomes so prevalent, that the bad usage becomes an industry standard.
To demonstrate, take an ASP executive, a traditional lawyer, and a contract negotiator--then lock them in a room... forever.
Just kidding. Instead, lock them in a room and ask them each to write down their definition of 'Force Majeure.' Force Majeure is a legal concept which can excuse performance under a contract in the face of disaster, riots, or acts of the public enemy.
To most business people, Force Majeure means an 'Act of God': fire, hurricane, tornado, or flood. Something unpredictable, devastating—an event which no business or individual could adequately prepare or continue performance in the face of. In the locked room, our lawyer probably derived this answer.
The ASP executive, if he or she has a technology or telecommunications background, probably believes that Force Majeure means anything that can’t be blamed on the ASP. This includes Acts of God, but also includes failures by third party suppliers, technology changes or delays in supporting product lines.
The negotiator takes a position between the two. Force Majeure equals whatever the two parties negotiate. If the terms are limited to Acts of God, fine. If third party suppliers can cause a suspension of the SLA, also fine. The negotiator doesn’t really care, as long as it is specified in the SLA.
In the ASP market, where most ASP deals involve some form of third party activity, this demands attention. The difference of opinion on this critical, but fundamental, point underscores the fact that most high technology ventures use the same language as the rest of the business world, but often with very different meanings. Customers dealing with an ASP may be surprised with the distinctions. These may even raise barriers to doing business. Potential ASP customers must pay attention to the realities defining their SLAs.
Earlier, we mentioned that SLAs are not a new concept. While true, the emphasis placed on them by the fledgling ASP business model is--unusual. In telecommunications and IT, SLAs emerged after the industries became more predictable. The SLA has a very different meaning in a field as new as ASP, where technologies, business models and players emerge, consolidate, succeed or disappear on a daily basis.
Most high technology purchasers whom we know have been ‘burned’ by at least one deal over the last two years. They feel cynical, often for good reason. This suspicion, in turn, leads customers to demand an incredibly detailed, powerful and punitive SLA before even considering their first ASP deal.
Press articles encourage this ‘scorched earth’ approach to relationship building by profiling tough SLAs forced upon vendors. They create more fear on the part of customers with 'train wreck' articles—our in-house term for an article about a big, failed IT project, painful to watch, but impossible not to look at--or one paragraph articles with advice on incredibly complex topics.
Investor-aimed advertisements add to the potential for misunderstanding and miscommunication. Huge ad campaigns by technology companies create an impression of the Internet as a stable, reliable and predictable entity; a system as reliable as the telephone. These ads may create a huge buzz among investors, but they also raise customer expectations.
A customer ploy that is a symptom of this cynicism is to bring a few provider advertisements to every negotiation. If the provider refuses to include something in a contract, claiming that it is not an industry standard, the customer places that provider’s latest ad--promising the exact same thing—squarely on the table.
Just so you know, this ploy rarely works—ads rarely reflect reality--but it sure is fun.
Each new customer horror story, hype episode generating analyst comment and advertising blitz raises the bar. Many of the new ASPs--including some huge industry players--attempt to meet these expectations and boost market share by way of seemingly ironclad SLAs. We say seemingly, because the overwhelming majority of these SLAs suffer from either 'Missing' or 'Manic' Terms.
Performance remedies and end-to-end guarantees qualify as the most common Missing Terms. A severe SLA that fails to include remedies--what happens if the vendor fails to perform--is not worth negotiating. Most ASPs realize this and include some form of remedy. Usually, those remedies aren’t tied to the business impact of failures, merely a mechanical formula or liquidated damage payment.
Customers who try to demand remedies that directly tie to lost business or opportunity costs are in for a long negotiation. This leads to the other common Missing Term, end-to-end guarantees. Most ASP deals involve at least one outside party, usually the telecommunications link between the ASP and the customer. We’ve seen ASPs that immediately agree to remedies, but cloud the SLA by suspending it in the face of failures by these third parties.
This leads us to the second malady, Manic Terms. An SLA is Manic when the service levels and remedies seem perfect, generous and look too good to be true. The question that enlightened buyers ask here is, "How can you stay in business if you do this for all of your customers?" Some ASP players started out in business-to-consumer eCommerce, where huge losses and fantastic guarantees were tools used to gain market share.
They now apply the same model to ASP deals. Some of the larger ASPs will do this until the eventual consolidation limits the number of players in the market. Smaller ASPs will do this out of desperation. The key danger with a fantastic, industry leading, cover-of-a-business-magazine SLA is that the company offering it will become so huge that they can afford to ignore it or so small that they will disappear. Either way, chances are that they will stop honoring the terms within twelve months.
A sales person will likely dispute every point we have made and will make in this series. They will insist that their SLA does not suffer from the faults we identify and that their ASP does not suffer from the same pressures bedeviling their competitors. Usually they tell the truth, as they know it. As customers, our job depends on determining whether the ASP knows what it is talking about—that is, that they are delivery focused—and can stay in business long enough to prove the model.
The IT industry has a history of over-promising and under-delivering. This history repeats itself in the ASP model, with promises of absolute (99.99999%) reliability, global availability and rock bottom costs. With this series, we identify critical weaknesses in the “guarantees” that accompany the ASP model, so customers can make an informed decision. We believe that the same factors that caused spectacular ERP debacles will lead to similar failures in ASP: