We sit down with Katie Madding (CEO, emhanceAI, ex-Adjust CPO) and John

Hobson (founded user research at Bungie and Blizzard, 20+

years on Halo, Destiny, WoW). The conversation walks through

two real case studies — a Skullgirls Mobile playtest and a

creative comparison — using emhance AI's facial-emotion

analysis (0.89 ROC AUC, basically "lie detector" precision).

The findings are uncomfortable. Skullgirls' tutorial pauses

combat to teach mechanics players don't need yet. Every

level ends with 5 passive screens stacked back-to-back. And

the creatives with "near-death experience" hooks within the

first 4 seconds genuinely outperform — not because Chinese

UA teams said so, but because the emotion data shows it.

If you build games or ads for mobile, this is required

viewing.

The link to the report is here: https://eu1.hubs.ly/H0v1sLM0

Peaksel case study: https://www.emhance.ai/success-stories/peaksel-emhance-case-study

⏱️ TIMESTAMPS

00:00 Cold open — designing the engagement curve

01:30 Welcome + Katie and John intros

03:48 How AI facial-emotion playtesting actually works

07:57 Skullgirls case study: the first 15 minutes broken

11:25 The attention budget — what it is, why it caps at 3

17:14 High vs low engagement moments in Skullgirls

24:13 Creative case study: Cat Match Two vs Sheep Swipe One

29:52 Why "near-death experience" creatives win — confirmed

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

📌 KEY TAKEAWAYS

— emhance AI uses facial-emotion analysis (0.89 ROC AUC, "lie

detector" precision) to identify exact moments of high

and low engagement during playtest sessions. Engagement

= emotion + attention combined.

— Skullgirls Mobile tutorial findings: pausing combat to

teach optional mechanics players don't need creates

frustration. Players come away thinking the game is

shallow because the tutorial fights are too easy and the

text walls killed momentum.

— John's "attention budget" rule: teach maximum 3 things

at a time, then give players space to digest. From his

Halo days: pop up a tooltip → if used, snooze 5 minutes

→ if not, repeat. Snooze 15 minutes after second use.

— The "5-stack of boring screens" problem is universal.

Win → victory animation → reward screen → loading

screen → cut scene → talking head → finally back to

gameplay. Nobody designed this stack on purpose. Real

playtesting reveals it.

— Why 6 out of 7 AI ads still suck: the concept is right

("save the character") but the small execution details

matter enormously. Emotion data lets you identify which

4 of 15 things in a creative actually worked rather than

throwing out the whole concept.

━━━━━━━━━━━━━━━━━━━━━━━━━━━━

🎙️ HOSTS

Matej Lančarič — User Acquisition consultant

Jakub Remiar — Game Design consultant

Felix Braberg — Ad Monetization consultant

🎤 GUESTS

Katie Madding — CEO, emhance AI (ex-CPO, Adjust)

John Hobson — Game user researcher, ex-Bungie/Blizzard,

20+ years on Halo, Destiny, WoW

This is no BS gaming podcast 2.5 gamers session. Sharing actionable insights, dropping knowledge from our day-to-day User Acquisition, Game Design, and Ad monetization jobs. We are definitely not discussing the latest industry news, but having so much fun! Let’s not forget this is a 4 a.m. conference discussion vibe, so let's not take it too seriously.

Panelists: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Jakub Remia⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠r,⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Felix Braberg, Matej Lancaric⁠

Join our slack channel here: https://join.slack.com/t/two-and-half-gamers/shared_invite/zt-3bckldvr8-8PXvzciMWdheOzED9hq0SA

---------------------------------------

Matej Lancaric

User Acquisition & Creatives Consultant

⁠https://lancaric.me

Felix Braberg

Ad monetization consultant

⁠https://www.felixbraberg.com

Jakub Remiar

Game design consultant

⁠https://www.linkedin.com/in/jakubremiar