The nine-page document (PDF) is titled PC Refresh and Windows 7 Migration Cost Analysis. Its scope is to "present a clearer picture of an enterprise migration process from the costs perspective, and suggest possible ways for cost reduction."
It discusses the migration process under the categories preparation and backup; role-based application install; application and system configuration; user re-training; self support/informal training; user downtime; post implementation support; domain settings and policies; and user data and personalisation loss.
Using tables, the document compares the costs for each one of these aspects for a traditional migration and one assisted by Zinstall; the time taken by the latter are lower and hence a cost savings is shown. Exactly how this is happening is unclear.
But when the document discusses hardware upgrades it omits some very vital aspects.
It talks of hardware upgrades and points out that in these days of tight budgets, upgrades, not replacements, are often resorted to, with a memory upgrade being the most common. But it does not mention the fact that a 32-bit version of Windows can only use up to 4GB of memory. This is a limitation in the Windows kernel.
Windows XP performs at an optimal level with 4GB of memory, even though it cannot use all of this memory due to the limitation cited above. Windows 7 requires 8GB of memory to perform at a similar level.
Upgrading the memory on a machine to anything over this, is useless if one is continuing to use 32-bit Windows - and to move to the 64-bit version would also require a processor and motherboard upgrade, given that there is no guarantee that the existing machine already has a 64-bit processor.
The document does not even canvass this possibility.
In offices, for the most part, hi-res graphics are not a priority; thus many motherboards in corporate PCs are cheaper all-in-one boards which have just two banks for memory. This means that one will have to replace all the memory in order to effect an upgrade; one cannot add to the existing sticks. This is not discussed.
In at least one place, the document states the bleeding obvious, which makes one wonder about the technical nous of the author(s). For instance, it says "When planning a migration project, it is important to consider the cost implications of upgrade. Apart from the cost of parts, two main aspects become technician time spent working on the upgrade, and user downtime during the process. It is recommended to perform this upgrade immediately before starting the overall Windows 7 migration process of the workstation, thereby combining the user downtime."
But it should be apparent to anyone, except the totally blind, that a new hard drive has to be installed immediately before the migration process is begun. Why would one interrupt any user to carry out the installation of a hard drive unless one is planning to do the migration of operating system as well?
The references provided at the end of the nine-page document are quite dated. Not a single one of the external documents cited is dated anything later than May 2010. And that is a long time ago as far as anything to do with technology goes.
Additionally, one has to pay hundreds of dollars to access the documents linked to in the document; evaluating their relevance is thus impossible.