Ugo Bui-Xuan

Director of Research @PlaytestCloud | Prev. UR Manager @Gameloft | ex-@PlayerResearch ex- @Ubisoft | Msc in Neuroscience. I love games, tech and cats

[Ugo Bui-Xuan @UgoBuiXuan]

@GramsRob @Femanatee Oui ça je suis d’accord. Classe moyenne et bobo c’est trop large, pas clair, pas défini. Mais je trouvais ton premier fil un peu violent alors que tes points sont bels et bien abordés. C’est juste superficiel car vidéo de vulga.
Mais quelques angles morts c’est vrai

[Ugo Bui-Xuan @UgoBuiXuan]

@GramsRob @Femanatee Le cadre qui a acheté à pantin c’est pour y vivre, pas un investissement locatif. Il ne pouvait simplement pas acheter à qualité égale dans le 11, le 20, ou le 19. Mais devant/voulant rester près de paris il achète à pantin. Il est acteur ET victime d’un même fait social non?

[Ugo Bui-Xuan @UgoBuiXuan]

@GramsRob @Femanatee Pour leur défense, Ils parlent bel et bien de la loc et du non encadrement des loyers..et la gentrification n’est pas que liée à la location mais aux prix des logements en vente

[Ugo Bui-Xuan @UgoBuiXuan]

@Steve_Bromley Love this

[Ugo Bui-Xuan @UgoBuiXuan]

@Steve_Bromley From the top of my head:
- The team not being sure about the audience even years in development
- not testing when the game should be
- big issues still in on release despite known for years
- fearing one tiny mistake (it happens!) could undermine months/years of trust building

[Ugo Bui-Xuan @UgoBuiXuan]

@Steve_Bromley Real question is what's NOT frustrating amaright? jk

[Ugo Bui-Xuan @UgoBuiXuan]

@Kirk_du_soleil I tend to agree with this BUT it really depends of the rest of your scale.
Do you use absolute (high/low) or relative (too high/too low) or something else? any examples?
Also by "lower" do you mean pure metric value or quality (ie not as good as they say)?

[Ugo Bui-Xuan @UgoBuiXuan]

@Steve_Bromley I usually run appreciation tests with 50 players (usually 25x2 profiles instead of 30x2… don’t ask why) and it’s enough.
A couple times I’ve actually compared data of 400+ live survey respondents and 50 playtesters survey and no statistically significant difference was found.

[Ugo Bui-Xuan @UgoBuiXuan]

@Steve_Bromley Really curious of where this number comes from. It might be an over simplification but I was taught that for hedonic/subjective data, 30 participants per profile was the minimum. With this, if distribution is normal you can run parametric tests. Am I missing something?

[Ugo Bui-Xuan @UgoBuiXuan]

@Kirk_du_soleil Yep, not comfortable either about having different rules depending on candidate desirability… but it’s a tough space and good/experienced UR are hard to get.
As HM we don’t always have the upper hand :/

[Ugo Bui-Xuan @UgoBuiXuan]

@Kirk_du_soleil He offered to present pieces of work he’s done in the past instead. I also asked him to send over or present the test he’s done in order to get his current job.

I know it’s risky but how do you do with sought after profiles that are in better negotiating position than juniors?

[Ugo Bui-Xuan @UgoBuiXuan]

@Kirk_du_soleil Fully agree with all your points.
I’m currently encountering a puzzling case: mid/senior UR candidate already in hiring process with a couple other companies warns he’d rather avoid test because he’s already busy with current job + other hiring/tests… what shall I do?

[Ugo Bui-Xuan @UgoBuiXuan]

@Steve_Bromley Such a shame, sounds like a great talk idea and I'd love to hear it somewhere else :)

[Ugo Bui-Xuan @UgoBuiXuan]

@DanaENight ...while some like prefer full support with full results presentation, ask for design/fix recommendations, sometimes include us in brainstorm meetings, etc.

No right or wrong, it depends on people/culture/process and it's always fascinating

[Ugo Bui-Xuan @UgoBuiXuan]

@DanaENight I'm curious cos UR culture and process can be vastly different from team to team, studio to studio. Even here at Gameloft we dont necessarily have the same process, depending on the dev team. Some welcome the data/insights but want to brainstorm separately amongst themselves...

[Ugo Bui-Xuan @UgoBuiXuan]

@darealsimmshady What the hell have you done this time, Sara

[Ugo Bui-Xuan @UgoBuiXuan]

@DanaENight And finally, internally did you separate "pure" UR feedback (from UR team) from expert feedback from consultants/journalists/etc?

[Ugo Bui-Xuan @UgoBuiXuan]

@DanaENight Hi I'm back with more questions! I finally watched the whole thing, what an amazing talk. Thanks for sharing this whole process!
On UR aspects, I'm curious: did UR present the results or just sent reports? Also did they make any recommendations, and were they part of discussions?

[Ugo Bui-Xuan @UgoBuiXuan]

@DanaENight (or maybe some answers are already in this talk... haven't had the time to watch it yet)

[Ugo Bui-Xuan @UgoBuiXuan]

@DanaENight Thanks a lot for the reply! Would love to know which was used for what type of research/playtest, pros & cons, how often, how it worked w/ development timeline and stage gate process...and what you would do differently if you could. maybe another talk :p