Reading Notes

Jeff Kaplan on World of Warcraft, Overwatch, Blizzard, and Future of Gaming

Episode: Lex Fridman Podcast

Notes — Jeff Kaplan on World of Warcraft, Overwatch, Blizzard, and Future of Gaming

Source: Lex Fridman Podcast #493, 2025. Jeff Kaplan is a former game director at Blizzard, responsible for World of Warcraft’s early expansions (Burning Crusade, Wrath of the Lich King) and Overwatch (2016).

Four questions [Adler frame]

Q1 — What is it about? A career retrospective covering the design philosophy behind WoW and Overwatch, the catastrophic failure of Project Titan ($83 million over seven years), how constraints produced Overwatch in six weeks, and what makes games feel loved versus merely shipped. Also an honest account of why Kaplan left Blizzard, the damage of online toxicity to creators, and his emerging thesis on the future of gaming.

Q2 — How is it argued? Through narrative and case study: Kaplan tells specific stories (the 3am GM call about Stranglethorn Vale, the Titan editor crash logs, the six-week Overwatch pitch) that illustrate general principles. The argument proceeds inductively — from concrete decisions to reusable frameworks (idea vs. vision, push/pull leadership, hotfix architecture).

Q3 — Is it true? The accounts are first-person and match known history. The frameworks (idea vs. vision, anticipatory hiring as failure mode) are useful generalisations from the Titan/Overwatch comparison, though the two cases are different enough that the comparison is illustrative rather than controlled. The toxicity argument is impressionistic but the Jay Wilson example is well-documented.

Q4 — What of it? The Titan/Overwatch contrast is one of the clearest documented cases of what overstaffing without vision produces versus what constraint and focus produce. The ‘idea vs. vision’ framework applies beyond game design to any product development context where a team must convert a good idea into a shippable thing.

Glossary

Idea vs. vision — Kaplan’s core distinction. Ideas are infinite and cheap (‘a secret agent who runs a flower shop’). Vision is the ability to shepherd an idea into existence: it requires technological plan + design plan + art style + pragmatic production reality. Titan had great ideas and no vision. Overwatch had a clear vision from week one.

Anticipatory hiring — Hiring people before you know what you’re building, expecting to find work for them. Kaplan considers this the cardinal failure mode of Titan: 70 environmental artists hired before the art style existed. Compare with Kaplan’s rule: ‘I won’t hire anybody on any team until we’re feeling like we gotta work overtime.’

Hotfix architecture — Engineering the client-server split so that server-side patches can be deployed without requiring client updates or downtime. Kaplan argues this is a core commitment at the architecture design stage, not an afterthought. For Overwatch, turning around a hotfix in 30 minutes was the target for critical hero exploits.

Push/pull leadership — The two modes of creative leadership. Pushing: when the team is too constrained, push them towards inspiration and vision. Pulling: when the team has too many ideas, pull them towards shipping. ‘The best feature we can add for the player is shipping.’ The leader is always in the opposite mode from the team.

Blizzard polish — The studio’s reputation for finished, responsive, craft-forward games. Kaplan’s account of what produces it: passionate QA embedded with dev teams (not siloed), direct communication channels between QA and developers regardless of hierarchy, hotfix architecture enabling fast response, and a culture of treating balance imbalances as bugs.

Online toxicity as design problem — Kaplan’s observation that social media incentivises maximum hyperbole (love it or call it a crime against humanity); neutral takes get no engagement. This self-reinforcing dynamic silences creators and normalises destructive criticism. His reference case: Jay Wilson (Diablo III director) retired from game development for ~15 years due to toxic feedback.

Section notes

Idea vs. vision: the Titan diagnosis

The failure of Titan was ‘a multifaceted failure’ — art, engineering, and design all failed simultaneously. But the root cause, Kaplan argues, was the absence of vision: ‘the hubris of Blizzard in that era’ where success with WoW produced the belief that the team could figure it out as they went.

The Titan engine (TED — Titan Editor) had 50% uptime in its final days: in a 40-hour week, the editor was usable for 20 hours. Kaplan’s frame: ‘Imagine having fighter pilots but we don’t let them fly.’ You can hire the best talent in the world, but if the infrastructure doesn’t work and the vision doesn’t exist, you have brilliant people doing busy work.

The art failure was not the quality of individual assets — Kaplan says ‘some of the best art ever made at Blizzard was made for Titan.’ The failure was cohesion: ‘the art looked like it could’ve come from 10 different games.’ Cohesion is a function of a defined art style, which is a function of a defined world, which requires a defined vision. None of these existed.

The six-week constraint and why Overwatch won

After Titan was cancelled, Morhaime gave the remaining ~40-person team six weeks and two criteria: ship in two years; have WoW-like revenue potential. Kaplan threw out the revenue criterion immediately (‘all fucking Monopoly money to me’) and focused on what could realistically ship.

Three two-week pitches: StarCraft Frontiers MMO (Indiana Jones in space, prospector vs. alien monsters — ‘here’s my money, sign me up’ but 5+ years minimum); Crossworlds (new IP); Overwatch.

Overwatch won because it was the only idea that was also a vision: clear art style (the Pixar aesthetic chosen early), workable hero roster, functional multiplayer prototype. The constraint of the six-week pitch — the team had to be all in for each two-week sprint and throw out everything at the end — produced a kind of rapid selection pressure that Titan never had.

Kaplan on what Titan taught: ‘You have to have the smallest group possible and try to get the idea across with whatever technology you can get your hands on… prove out that idea, and once you know what you’re doing, then you expand the team.‘

Blizzard polish: what actually produces it

Two structural factors Kaplan highlights:

  1. QA integration: At the peak of Blizzard polish, QA members sat with development teams, had direct communication lines to developers regardless of hierarchy, and included specialists (e.g., a QA member who could hit sniper shots in a shooter reliably and could detect a single frame of input lag). The relationship was built rather than assumed. Kaplan gave QA members his home number: ‘If you think we’re gonna get raked over the coals on this, call me.’

  2. Hotfix architecture: Designing the client-server split so the server can be patched without client downtime. This is an architectural decision at the foundation of the engine, not something you can retrofit. It allowed Overwatch to disable a hero with an exploit within 30 minutes. Without hotfix architecture, every balance issue becomes a multi-day patch cycle.

The feel of polish — ‘you can feel the love and the craftsmanship of the developer’ — is not just aesthetics. It is responsiveness: ‘Do they clean up their fucking yard?‘

Online toxicity and the creator extinction risk

Kaplan’s Van Gogh thought experiment: ‘Would Van Gogh have existed if Reddit and all these things were out there?’ The historical question of how many people could reach a creator directly in their lifetime (hundreds at most) versus now (anyone with 10 devices) represents a qualitative change in the feedback environment.

His concern is not that creators are criticised but that the volume and valence of feedback creates selection pressure against the vulnerable, who are often the most original. Jay Wilson (Diablo III) is his example: ‘We lost essentially… a brilliant design mind for 10-15 years, and it feels criminal to me.’

Kaplan is careful not to romanticise the pre-internet era — he acknowledges the early WoW forums shaped online culture in ways he now regrets — but the design of social media (upvote/engagement systems that reward extreme reactions) is the structural problem.

Why Jeff left Blizzard

The immediate trigger was Overwatch League and its sequelae. Kaplan had co-championed the league concept (regional teams, player salaries and protections, esports professionalism) but the execution became a financial vehicle. A media rights deal with Twitch created technical and content obligations that consumed the Overwatch team’s capacity, at the expense of the live game (new heroes, maps, world events) and the original Overwatch 2 vision (PvE story content).

The structural problem: the Overwatch League business model (in-person ticket sales, merchandise, eventually media rights) was oversold to team investors (billionaires who had been told the League would rival the NFL). When that collapsed, investors defaulted back to pressure on the live game for in-game monetisation. ‘We had all these people who basically invested in the game, and then they started to express their opinions.’

Kaplan on the departure: ‘I believed in Overwatch 2. I think we could have made a great game. I have a lot of hindsight of, like, how I would have designed that game differently with what I know now.‘

Cross-references