For most of the last year, we have been rebuilding our corporate network infrastructure to meet future performance and security requirements. Making use of features and capabilities barely imagined five years ago should let us meet our connectivity needs for the next several years.
Now, however, we have a challenge. A great team of consultants did the design and installation work, but the future operation and evolution of the network will have to be managed by our own employees--and they don't have the appropriate experience and knowledge yet.
Our engineers are quite capable of learning the new design, and managing and monitoring platforms and feature sets. What concerns us, though, is how we will fit all the necessary training and knowledge transfer into a schedule that is already packed with everyday support work and business-critical projects.
So, in a world where we refresh our core technology every three to five years, how much should we be investing in training our support staff? And as we look to save money and preserve our options for future growth, how can we afford the investment in human capital that this training requirement implies we should make?
One way to address the "how much training" question is to look at how much our vendors train their support engineers. Although it is likely to be more than we need (we buy support contracts for the hard parts for a reason), it can give us an upper boundary to work with.
Surprisingly, when I canvassed our main partners, I got a wide range of answers. However, I came up with illustrative ranges:
â¢ New engineers get around 10 weeks of training in their first two years and between four and eight weeks after that. They get supervision for the first year and are usually not permitted to do "lead" work until their third year. Most employers also require new engineers to get appropriate certifications.
â¢ Moderately experienced engineers (who are generally required to have and maintain appropriate certifications) get an average of six weeks' technical training each year, plus one or two weeks of nontechnical training. Many vendors are moving to self-study and distance learning approaches, but everyone still uses some classroom time.
â¢ Subject experts spend the most time in class--an average of 12 weeks per year--but at least one-third of that time is spent teaching.
Most vendors said they aim to cross-train their support engineers on multiple products, but are not always able to do so when new products launch. We have to train our people on several different vendors' products and how they integrate. Still, we don't use every product from every one of our vendors, so our training universe is smaller.
We probably need an intensive training program for the first few months of a new deployment, followed by a few weeks per year of refresher training. I would like a model that does training one day a week and lets the staff put what they learn into immediate use, but I don't know if that is possible.
So, can we afford this? A better question is, How can we not afford it? We risk wasting our investment in technology if we don't invest in our people on a systematic basis. If we can't support what we deploy, we will need help from third parties--and, in the long run, that's even more expensive.