of time then we either need super-human staff or more of them (see previous section).
There are a few other initiatives that can be taken to make the best possible use of time:
Multiple shifts —depending on your team availability and that of your test environments, you may be able to steal a march on your deadline by requesting that your team go to multiple shifts (and providing they agree).
This is where some of your team, rather than work standard office hours, are scheduled for the duration of the test project (in full or in part) to work in the evenings, eg. 4:00 p.m., to midnight. If you’re really up against it, you might even consider the ‘graveyard’ shift—midnight to 8:00 a.m., or similar.
‘Round the clock testing’ can provide you with significant gains in time however there are a few ‘gotchas’ to watch for:
- Be sure you don’t burn your team out, they’ll be of no use to anyone further down the track if you charge too early. The key is to plan and pace, plan and pace. Make sure that you’re always revisiting your plan, as a project manager (which is in essence what you are in this role—with a twist, of course) you should be doing this as a matter of course anyway. Ensure that the intense effort is going in where it has the best effect.
- There’s no point in multiple shifts if any defects that are logged cannot be attended to until the next day. One of the keys to testing is the timely turnaround of defect fixes and if you’re at the stage of multiple shift testing then I’d suggest that leaving high-priority defects longer than a few hours before they’re at least investigated is going to cause you problems. This does mean however that you will need to get your support/development team to buy in to this idea as well, which might not always be easy. You may also have to accept the fact that you may have impassioned your test team into these crazy working hours but the development team might not see things quite the same.
- Make sure that by going to multiple shifts, that you are not merely doing after hours what could be done inside hours with better utilization of resources eg. test environments (refer to the discussion on duplicate test environments). I once had a test manager who went to a second shift and all his second team did was perform testing that could have been done inside hours with a second test environment. This may be less costly however a problem usually has more than one solution; the trick is to choose the best.
- Ensure that the effort of the additional shifts is monitored and the output quality checked. Unless you’re a total insomniac, you are probably not going to be able to manage this hands-on yourself. It might be worth your while appointing the more senior members of your shift teams as shift leaders and request a report each day of the previous night’s activities and status.
Appropriate scheduling —I think we all tend to put down eight hours (or so) per day for testing just as we do for any other task. However reality dictates that you’re probably going to get no more than six (or so) productive hours on average per day. Once you have allowed for meetings, interruptions, down-time, paperwork, sick leave etc., even six might be ambitious.
There’s also the question of ensuring that you have allocated the appropriate team members to the tasks that they’re going to be the most