[Testing] Careers In Games QA Management

So, you want a career in video game quality assurance? Please allow me to offer one word of advice...

NO!

Outside of the video game industry, quality assurance can actually be a viable career option. Experienced testers are in high demand, and people who can think logically and can break down tasks into sections small enough to be properly tested will always have a job open for them.

However, inside the video game industry, testers are considered fodder for the machine. Long hours, low pay, an average "career" length of five months...it adds up. The only way that a career in games QA is going to go anywhere is if you get into lead-level or above QA, and that's where this article comes in.

In order to make it in QA management, you not only have to have the basic skillset of any tester, but you also need to have at least some knowledge of how the business of the games indsutry works, have some understanding of development, and be able to negotiate and understand the consequences of your negotiations.

For the business side of things, you need to be able to think about the ramifications of missing your ship dates, the tradeoffs made between bringing on additional headcount versus the potential savings for technical support calls, whether shifted employees or overtime will be better, what tasks should be outsourced versus what people want to be outsourced, etc.

We'll use outsourcing as an example here. Right now, a major effort towards outsourcing QA is in place in this industry. That said, unless a task is atomic enough that it can be handed off to someone new, such a task is ill-suited towards outsourcing. Configuration testing is great to outsource. You provide a list of configurations, a build, and a list of test cases. You'll pay on average $100 to $200 per configuration for testing, but in the end, the money spent will be saved on support calls. Configuration testing works because the amount of work is finite, creativity in testing isn't required, and spot-checks can be made based on configs in house.

Understanding development is a major necessity to break into QA management as well. You need to keep up on the tech, development methodologies, stages of a product's lifecycle, etc. These things help you understand when to start up certain tasks, when to pressure for certain tasks on the dev side to be finished (for example, a build for the config labs so there will be enough time to fix config bugs), etc.

Finally, we've got negotiation. Negotiation is the fine art of letting someone else have your way, and it is a critical skill if you want to make sure that the major bugs that are in your bug database are going to get fixed. This, combined with understanding the development process, is the major key to success in QA management.

For example, let's say you are having a bug committee meeting. (Some companies call this triage.) In a bug committee meeting, everyone goes over the active bugs in the databases and people try to reach a consensus as to which bugs are going to get fixed pre-ship, which will be fixed in a patch, and which are going to be dismissed outright. These are common during the end stages of any game development project because people are trying to reduce the amount of asset and code changes to a bare minimum.

Let's say that you've working on a multiplayer FPS with a zone-control mode similar to Halo's "King of the Hill" mode. There are two bugs currently being considered by triage. One bug is that if a player shoots 10,000 bullets in a row and doesn't hit a single enemy, that player's client crashes to the desktop. The other bug is that 5% of the time, the HUD element that tells you that the zone is being controlled by the enemy reports incorrect information. The lead developer on the project says that he only has enough time left on the project to fix one of the two.

Do you push for the crash bug fix, even though the likelihood of it ever being encountered in the real world is slim to nil? Do you shoot for the HUD item being fixed because it will be encountered by everyone who plays multiplayer, even though it isn't a stability issue? Do you offer up some lower-priority issues that are still open for the chopping block to make the time available for both to get fixed? What do you do?

So what will you get if you manage to make it to the level of test lead? On average, test leads make between $30,000 and $35,000 per year in the United States. Test managers start at about $50,000 and go up based on experience. Given that the average game tester makes between $17,000 and $20,000 per year, the pay increase is generally worth it. The only downside is that once you go above tester, you're usually salaried, so don't expect any overtime. Most companies that have salaried employees also have profit sharing or bonuses for said employees, so make sure that you ask about it before signing on the dotted line.