Bad Decisions Podcast

The Varun interview yields a handful of genuinely operational insights - iterative trailer development as market research, the artist-vs-engineer phenotype split in game dev, photogrammetry logistics, the aura farming mechanic as a designed virality system - but a 119-minute runtime is heavily padded with host affirmations, a 20-minute sponsored Autodesk Flow Studio walkthrough, trailer reaction banter, and recycled AI discourse, dragging the ideas-per-minute rate well below average.

“we are the weirdest game devs because we've been shipping this totally startup style”

“Somewhere between 50 to 75% of a AAA game budget goes into marketing”

There are a few genuinely non-obvious framings - applying YouTube retention and dopamine thinking to game design, the 'if you make a third-person game there is no indie game' observation, and the Pygmalion effect applied to public building - but the episode spends considerable time on well-worn takes: creators will own all industries, AI is a tool not a replacement, streaming will replace ownership (music did it so games will too).

“if you make a third person game, there's no such thing as an indie game even if you're independent. Because any mistake you make, any small gap that's there between you and the billion dollar franchisees are just going to...people are just going to call it out”

“I feel like you think about it the same way you do as of a video, which is like how do you have maximum retention time? How do you also have clips that go viral?”

Varun Mayya is a legitimate operating practitioner - he runs a 560-person content company with real enterprise clients, self-funds a game studio, and is actively doing the thing he describes, not theorising about it; he is not a career thought-leader. However, his domain is niche (Indian indie game dev + creator economy) and his seniority in a global B2B context is modest.

“We're now 560 people on the content side of the company”

“we looked for Unreal Engine VFX artists in India. We could only find two or three in the entire country”

Varun supplies a respectable number of concrete specifics - team sizes, shoot logistics, view counts, marketing budget percentages, asset volumes, and timeline estimates - which is above average for a creator-oriented podcast; what keeps the score from being higher is the complete absence of hard financials for the game, vague release windows, and a sponsored product segment that is definitionally abstracted.

“we sent I think a ten man team there. It was like a nine day shoot...we have like a thousand assets. Right. Out of which maybe like only like 100 were unusable”

“That was the highlight that got like four and a half million views, uh, on IGN China”

The hosts ask mostly open, narrative-inviting questions and respond with near-constant affirmation ('Mhm', 'Yeah', 'Right'), rarely probing contradictions or pushing for harder numbers; a few specific follow-ups (budget cap question, asking about mocap actor sourcing) surface but are exceptions, and large claims like '60-70% chance of making a good game' or 'millions in free marketing' go completely unchallenged.

“So why don't you just tell us a little bit about how the past few months have been for you guys at AOS Games?”

“Do you have a max budget in your head that you would say that I'm not going to pass this amount no matter what?”