Originally appeared in MSI Magazine
January 1, 2000

Rooms for Rent

It’s funny how history often goes in cycles. Over the past few years, we’ve seen a number of hallowed and discredited ideas alike make their comebacks. Remember the network computer or computer-aided software engineering? Today, of course, it’s more fashionable to use the terms thin clients or design and analysis to describe these concepts. What’s refreshing about the IT industry is patience—or shall we say desperation—of users not to give up on technologies during the first go-round. Otherwise, what else would explain our patience with Microsoft, which of course always gets it right by version 3.

And so it seems like many of us about to give outsourcing—or was that service bureaus or time-sharing—yet another try. Few companies had complaints with service bureaus, which handled eliminated the headaches of running peripheral functions such as payroll. But, by focusing on generic functions, service bureaus remained limited in scope.

Outsourcing broadened that idea, chalking up some notable successes. But there were also the horror stories, like locking some customers into 5 – 10 year contracts with obsolete technologies.

To make outsourcing more palatable, we’ve given it yet another name: application hosting. In place of the outsourcing vendor, we’ve retitled them application service providers.

Raising an old idea from the dead doesn’t necessarily mean that we’ve run out of better ideas or that the notion was bad to begin with. As most of us are all too aware, the pace of change is accelerating. Manufactories are no strangers to change, having been under the gun to rev up product cycles over the past 20 years, from just-in-time to Internet time.

New rules of competition demand new approaches, while new technologies open new possibilities. Sometimes, those changes take us back to old ideas whose times have finally come.

Changes in technology and the rules of competition are finally making a compelling case for application hosting. Some of the changes, such as the emergence of the Internet and web computing are obvious. Web computing offers ASPs economies of scale by eliminating deployment costs, and if the Internet ever becomes secure enough, could put a lid on networking costs. Furthermore, changes in the competitive climate are driving second and third tier manufacturers to adopt the same sophisticated supply chain management practices as their trading partners, is opening a potential new mass market for ASPs.

The result is the prospect that small and midsize manufacturers can now budget enterprise systems at a predictable monthly cost of $300 – 600 per user. ASPs are offering promises to get customers into production between 3 – 12 months, with service agreements stretching from 2 – 5 years. That sounds a lot better than spending six or seven figures for a permanent solution that won’t deliver results for 12 – 36 months.

However, just because the stars appear to be aligning for ASPs, business practices for third party hosting remain all over the map. For instance, while the emergence of ASPs should open the way for manufacturers to rent applications, vendors such as SAP are still insisting that customers buy licenses, regardless of who runs them.

What types of providers are best qualified to become ASPs? Phone companies, a consulting firms, software vendors, or middlemen who rent capacity from somebody else? At this point, everybody is trying to horn in on the act.

What about economies of scale? Can ASPs deliver cheap service by allowing its customers to have all kinds of configurations or customized extensions of applications? If they charge extra for “changes,” what constitutes a chargeable change? And, if manufacturers decide to buy basic ASP services, can they still remain their competitive edge with plain vanilla software?

The upside of application hosting is that they can be structured to resemble capital equipment leases with fixed, 3 – 5 year expiration dates. The upside is that at the end of the agreement, your company’s options are open. If the business changes or the ASPs service falls below par, you’ve got a great escape clause. The downside, however, is that while your options might be open, so too are the cost estimates for migrating to a totally new solution.