In Johanna's last column, she discussed ways test managers can measure their output in working terms of problems solved, crises averted, etc. Prompted by a comment from our sticky-minded audience, this week Johanna shares some ways test managers can assess their performance against specific management deliverables likely to be high on an organization's priority list.
In my most recent StickyMinds column, I discussed how to measure some of a test manager's output. I received some very thoughtful replies, one of which I'm going to discuss in more detail here.
Surjya Mohanty asked what the test manager's measures are from the organization's perspective.
In short, the test manager owes three specific deliverables to the organization: portfolio (multiple project) planning, hiring strategy and planning, and a staff retention strategy. Within those deliverables comes a number of the ideas you folks commented on in my last column: the ability to estimate project duration, coaching and mentoring staff, understanding what's done and not done, etc.
Now maybe some of you are saying, "Hmm, I don't have access to the project list, so how the heck am I going to manage my project portfolio? What about process improvement? Don't I have some responsibility there? Sure, I'd develop a hiring strategy if I had open requisitions." My advice is that even when you're not sure you have the authority, you should still address the issues of portfolio management, hiring strategy, and staff retention.
Managing the Project Portfolio
As a test manager, I work hard with my project managers and development managers to understand what's in planning for development and what's currently in active development. Some organizations do portfolio management in a systematic fashion, but most organizations do three- to nine-month planning at best. Actually, if you're in a young organization, it may make a lot of sense to do continual portfolio planning, since you'll be listening very hard to your customers, and you'll change what you do frequently. For older organizations with already existing customers, the changes are less frequent, and more systematic portfolio planning is possible.
How do you assess how well you do portfolio planning? Here are some measures:
- Are you able to rank projects, and assign your best people to the most important projects? Especially as a test manager, you need to put the most valuable people on the most critical projects. Of course, you need to balance assignments with cross-department learning opportunities and other hiring and retention strategic measures, but putting your best people on the most important projects is one organizational measure of a great test manager.
- Are you able to influence the portfolio planning, so that projects are not started, stopped, started again, with the associated context switching?
- How many projects surprise you? I'm sure that we've all been surprised by projects we didn't know about in advance, but if that happens to you more than once or twice, it's time to investigate why that's happening. You are much more effective as a test manager if you can plan to use specific people on specific projects, not just robbing one project to satisfy another.
- Another measure is how often you need to move people from one project to another. The more often you move people, the more you need to look at your hiring practices. If you have a scarce resource, it's time to change job descriptions and hire differently.
Hiring Strategy and Planning
How many times do you hire someone just like the last person? Or have you hired someone based on a tool they know or their GPA? Test groups require a variety of people, in order to thoroughly test the software. And if you have a scarce resource, it's time to develop more of those capabilities in your group. Some measures I've used are as follows:
- Do we have people who can create test strategies, test plans, and test cases? Do we have test architects, and people who are expert






