I’ve been with the company seven months now, and I’m starting to wonder if I’m becoming the bottleneck on my team.
I’m the sole QA across several areas — slot games, mini-crypto games, and progressive market games. In that time I’ve released around five slot games and eight mini-crypto casino games, all of which have performed well.
The concern is that there’s a growing backlog on the mini-games side, and I can only realistically test and release two slot games a month. I’ve been using Codex to automate some of the HTML-based mini-game testing, which helps, but there’s still a significant manual workload that automation can’t cover.
Our dev-to-QA ratio is 5:1, and I’m starting to think it might be worth raising the idea of bringing on another QA. I just don’t know the best way to approach that conversation with my manager especially being relatively new.
Has anyone been in a similar position? How did you frame it did you go in with data, or just open it up as a question?
Personally I prefer something like 3:1 dev/QA, or even 2:1 for more complex or critical apps. 5:1 even for gaming seems excessive. What happens when you go on leave, do things just back up until you return?
You might find the manager is already aware of it, it won’t just be yourself that is watching the ever-increasing backlog. I don’t think there’s any benefit to working the numbers initially, just raise it with them as a query first. If he does need justification (and if he’s reporting to bean counters then that may be the case) then you can start number crunching and putting together a report. I’m not sure being relatively new has any bearing on whether you can highlight a potential bottleneck.
Funny enough I haven’t taken much days off this year yep. Only three days. I believe things get backed up. Planning to take some holiday in a few months
Perhaps there are more efficient processes to adopt? Ones that are more quality engineering and quality coaching focused, that bring in other people in the team?
You can only automate or AI stuff till a certain point after which you’d still need capacity.
Your right to PTOs should never be affected by your work load (at least we don’t do that here in Germany)
Asking for another QA trainee is the only way to go here. After some time, that should give you the capacity to move things to tooling and make the process faster.
I’m more of a software tester rather than a QA and I want as much of my time as possible being spent on the discovery, investigation and experimentation of risk as I can.
Ratio of 10 developers to me is fairly normal and I am rarely overloaded at those levels, with Ai advances that ratio may even go upwards.
This model works well when I am not taking on other activities that other roles are just as good at or often even more efficient at and also I am allowed to work on multiple products in parallel. The model does not work on single product allocation as there is often not a continous flow of things to test and you end up with moving yourself into other activity areas in the name of productivity.
James Bach wrote about the idea of Test Jumpers over a decade ago and it’s still my favourite software tester model, not QA, not QE, not SDET those are different roles that I can and do help out on those sort of activities but as a tester my focus remains on that discovery, investigation and experimentation of risk.
The challenge though is if developers have not been applying due diligence, for example them picking up that automated coverage you have picked up even when they would be more efficient at it and you end up doing their job plus your own you will become a bottleneck, this is the hardest bit to change and across the industry its often just accepted. This is where I’d start the conversation, design a model that really suits your strengths and then try and get the team on board whilst also discussing any gaps that can be shared and transferred.
Testers also shifting into build QE or process QA things that architects and PM’s also do is also common, there are some arguments for this and I’d say AI model changes will empower everyone on the team to be builders but I still want to hold on to my core software tester hat as long as I can.
I’d recommend against more testers particularly as you already have a very low developer to tester ratio.
That’s interesting thanks for the information. I also like reading James Bach and Michael Bolton newest software testing book. I do AI to help with tasks and some automation aspects too. I also enjoy the exploration and investigative side of testing too.
