Posted on 7 July 2008 by James Blake

Before I can tackle this myth, it is important to consider how this myth has come about.

While it is a natural assumption is that when you move technology outside of your organisation’s boundaries you will lose control.

This mindset is largely fuelled by a history of projects where third-parties have taken over responsibility for either the wholesale operational management of IT infrastructure (outsourcing) or specific problematic elements of IT infrastructure (managed service). Typically all outsourcing or managed services do is move the problem from one party to another – the infrastructure is typically made up of a series of technology solutions strapped together to try and solve a business need.

What is gained by using a third-party is specialist staff to run the infrastructure, allowing internal IT staff to concentrate on strategic issues. Another attractive draw is that the Service Level Agreements for service are also externalised to the third-party, the burden is removed from the IT department.

Software-as-a-Service, in comparison, is not about moving the problem from party A to party B. Software-as-a-Service is all about solving the business need in a different way, abstracting it away from the technology solution.

Unlike an outsourcing or managed service offering which maintains the status quo, albeit in an externally managed fashion, Software-as-a-Service seeks to provide a distinct solution to a business problem as a black box - the user need not worry about what hardware the service runs on; how replication works within the service; how to migrate data when the underlying storage platform goes end-of-life; or the structure of the relational database – they just care whether it provides their users with the solution they need; will it be available when I need it; how secure is my data; and how much does it cost?

Many managed service providers are now reinventing themselves as Software-as-a-Service companies.......without reinventing their offerings. They are built from the same collection of historical on-premises solutions, just accessed over the Internet with some form of reporting and provisioning overlay. Some customers are attracted to this, as the on-premises vendor’s names are familiar to them. The issue is that a managed solution posing as a Software-as-a-Service solution that is based on on-premises solutions has the same integration issues as you would have on premise with the added complication that you get the lowest-common-denominator of configurability and functionality, as you are trying to use a multitude on on-premise point products to solve differing issues for different customers.

At the same time Software-as-a-Service vendors have started to consolidate. This consolidation won’t solve the issue either as the consolidation results in two different storage platforms; two different configuration databases; two different billing systems and two different Service Level Agreements . We’ve been used to this problem for years in the on-premises marketplace for decades – you find a product that does something you really need, shortly before they are acquired by an industry behemoth.

On-premise vendors are also reinventing themselves a Software-as-a-Service offerings. In an environment such as email management this simply displaces a part of the problem from on-premises to the cloud – it moves the technology solution to the cloud, not the satisfaction of the business need.

Consider a typical infrastructure in the space Mimecast operates – email management. Rarely is a single point product used to manage email, it is normally a ecology of disperate point solutions – firewalls; Denial-of-Service prevention; intrusion detection systems; bridgehead servers for email routing; back-up Exchange server; encryption key servers; encryption gateways; archiving database servers; archiving reporting servers; anti-virus; anti-spam; anti-phishing; disclaimer management; attachment management;...... If each vendor embraces the Software-as-a-Service model, you’ve still got the same integration issue you’ve always had– multiple places to provision; multiple places to configure; multiple dashboards to watch and multiple reports to read.

A good Software-as-a-Service offering will be based on an infrastructure designed to offer scalability and maintain availability beyond that the customer is experiencing with their on-premises products. The back-end software will be multi-tenant aware from the ground-up, with each each component tightly integrating with to provide an end-to-end solution to the business need.

The business need Mimecast aims to solve is the management of email through its entire lifecycle – hygiene, policy, continuity, retention and discovery. The efficient management of email enables legislative compliance, internal governance, cost reduction, better protection from future litigation and enablement of a rich source of business intelligence – business needs. With good Software-as-a-Service offerings the IT department and especially the user should be totally abstracted from the technical solution and be free to reap the benefits of their newly resolved business issues.

Furthermore, Software-as-a-Service should be able to extend on-premises infrastructure not permanently strive to replace it. Some customers have long-term commitments to their previous on-premises infrastructure purchases. Software-as-a-Service vendors need to adapt to how the customer has chosen to run their environment, not force them to compromise. If you force the customer to compromise you’re moving from the business need to the technology solution again.

Mimecast, for instance, tightly integrates with a customer’s Microsoft Exchange environment, not only to retain internal emails and ingest historical email, but also to create a custom reputation database for each user by analysing historical communication patterns – before they even join the service. Mimecast also extends onto the user’s desktop, providing seamless access to email from within Microsoft Outlook during an outage, allowing users to perform self-service recovery of deleted email with a simple drag-and-drop and even perform searches through archived data – all from within the email client they use every day. Again, focus is on the business need – “our user’s want to use a single client for email and don’t want to swap to a web browser” - not the technical solution.

Far from losing control with Software-as-a-Service, when IT managers select an offering that provides exactly the solution they need which is built with scalability, availability, feature richness and granularity of configuration in mind, they can free themselves to concentrate on how to use these tools to implement best practice and gain competitive advantage. Instead of losing control, the IT manager now has a carrier-grade solution at his control without the need to design, deploy, manage and refresh it.

Software-as-a-Service has not yet reached the stage where the vendors for each solution fully integrate with one another – but this day is coming. But the level of abstraction Software-as-a-Service provides today, being able to pick one solution for each business problem rather than having to integrate multiple point solutions for each business problem, explains the wide-scale adoption we have seen over the past couple of years.

As long as IT managers avoid managed service and point solutions manifesting themselves as Software-as-a-Service, there should be no reason why IT managers can’t regain control of their ability to provide service to their internal users.

 

Category: Software-as-a-Service

Technorati: