Ok - caveats out of the way, let's get into the meat of it.

First of all, can we acknowledge that the question "Can we trust them?" is loaded?

It's binary, which makes it impossible to give a correct answer.  And besides, the question itself is just a proxy, a stand-in for a deeper inquiry about the potential success of the project.

If we instead view this problem on a spectrum—IE "What is your confidence level in your potential partner?"—you’ve now made an impossible problem workable.

After all, there are a million ways to due-diligence a studio and project team to increase your confidence in a project's potential success.  From in-person interviews, to speaking with past clients, to simply getting on a plane and visiting the studio.  It’s just a matter of effort.

Which is why the confidence level question should be considered alongside "What is the potential risk/downside if things go wrong?"

Are you asking someone to define your quality bar, or simply to deliver to an established standard?  Some projects have far more future dependencies on them than others, and as I’ve discussed before, you should invest more in higher leverage initiatives through a combination of more time spent in due diligence and paying for more experienced talent.

Of course, one of the easiest ways to gain confidence in a partner is to get a direct referral from someone you already trust.  This works great, and it’s one of the highest value signals you can have as to a potential partner's success.

But…and I’m going to admit I’m obviously biased here…I believe the amount of value placed on referral channels in games is excessive.  Too often I see companies exclusively use referral channels in their searches, ignoring the vast majority of the industry outside their network.  In fact, many consider “ask your friends” as THE prescribed method to find ExDev partners.

There are many reasons for this, but in my personal experience I’ve found I’ve personally overweighted referral channels when:
1) I didn't have a very strong understanding of my own requirements.
2) I didn't know how to perform comprehensive due-diligence because of #1.
3) I didn’t know where to find additional sources of potential partners, even if I wanted to consider additional candidates.

In these cases, by “trusting the referral to be good,” what I was actually doing was outsourcing the responsibility for my own project’s success. I was hoping that getting referred to a 'good partner' would solve my problem, rather than spending time defining the problem myself.

And guess what?  Failure!  Even though it was a trusted referral!

Once we were able to dial in exactly what we needed in a partner, we stopped soliciting referrals. We instead went out into the market and actively searched for studios with the experience we required.  And our quality went up, costs went down and our failure rate dropped to near zero.