The Zombie Stack Exchanges That Just Won't Die
Organizations have the option of building the expertise to navigate the intricacies of imaging and copying from extremely varied digital media or outsourcing these tasks to other professionals who are already experts (ie data recovery pros, forensics experts, digital video preservation services). Assume that any outsourcing would be within specific parameters assigned by the organization.
Courtney C. Mumma
Speaking as an individual who has a fair amount of content to convert (home movies and family photos), there are basically two paths for in-house conversion: professional equipment and minimalist equipment.
Professional equipment for converting some types of media (photos) can be extremely expensive, but professional equipment is also much better at handling large volumes.
Minimalist equipment is much less expensive and slower, and may also produce lower quality results (depending on the media type and investment).
Taking photos as an example, there are four types of solutions that I am familiar with:
a) desktop scanner - cost \$100-200
b) automatic feed/work group scanner - cost around \$500
c) DSLR and copy stand/table - cost \$1000 and up
d) Professional scanner - cost (I don't know myself, but it's in the \$10,000 or more range)
When I started to work on my own photo conversion project, I talked to a senior editor at a photography magazine and his professional advice was that solutions a and b would yield mediocre quality. On the other hand, solution c would give results that were moderately close to d. Solutions a and c are the slowest, b and d the fastest.
So when an organization is considering a conversion project, they first have to figure out the quality, budget and schedule constraints that they are working under. Then they have to determine if there are any special factors, such as proprietary material that you don't really want to have to ship out.
There are really too many possible scenarios to give any hard and fast rules as to whether it's better to do it in-house or contract. Organizations should do their research. Be realistic about the time, cost and inconvenience, and it should become clear which solution is the most cost effective.
The largest factor in deciding whether to outsource or to convert in-house is the scale of the project.
Outsourcing is a variable cost. If the project expands in scope, the costs will increase roughly linearly.
In-house digitization has a fixed cost (equipment) and variable cost (staff time). If the project expands in scope, the fixed cost per digitized object will decrease inversely while the variable cost will increase roughly linearly.
This breakdown holds true for other assets. For instance, outsourcing allows you direct access to expertise, at a price that scales with the size of the project. Inhouse builds the expertise of your staff. The larger the project the less the cost of this expertise, but also the greater time staff will have to devote to the project.
At a given project size, it makes sense to invest in the equipment and staff to keep a project in-house, but smaller projects will likely be more economical to outsource.
A couple of additional points to Nick's excellent answer...
Depending on the target of the data recovery, specialist equipment may well be required which may be excessively expensive and/or difficult to obtain due it to being obsolete or simply in short supply.
Utilising an external data recovery company means taking your data off site. This introduces some degree of risk. This approach may be impossible if you are dealing with sensitive data, or simply have an organisational policy which precludes taking data off site.