Curse Inc: The Gaming Empire That Amazon Had to Have

Picture this: A French teenager, so consumed by World of Warcraft that he plays 16 hours a day, gets kicked out of his parents' house for his gaming addiction. Most would see this as rock bottom. Hubert Thieblot saw it as the beginning of a $60 million empire.

The year was 2005, and while other 20-somethings were finishing college or starting conventional careers, Thieblot was channeling his obsession into something extraordinary. His interest in World of Warcraft led him to create a website to organize and share modifications, add-ons, and plugins to change player experience via the user interface. What started as a simple mod repository would eventually catch the eye of tech giants Amazon and Riot Games, reshaping how millions of gamers connect and play together.

I. Introduction & Cold Open

The gaming world stood at attention on August 16, 2016. Amazon.com Inc. announced via subsidiary Twitch Interactive that it would acquire Curse, Inc. for an undisclosed amount—though industry insiders whispered numbers well north of the $58 million in funding from investors including GGV Capital, Idinvest Partners, Riot Games, SoftTech VC and Ventech that Curse had raised over its decade-long journey.

This wasn't just another tech acquisition. This was Amazon, through its billion-dollar gaming platform Twitch, acknowledging that the future of gaming wasn't just about watching streams—it was about the entire ecosystem of tools, communities, and connections that make gaming culture thrive. The hook that made this story remarkable wasn't the exit; it was the journey of how a hardcore, 16 hour-a-day addiction to World of Warcraft transformed into one of gaming's most strategic acquisitions.

What makes the Curse story particularly compelling is its timing. Founded in 2006, just as Web 2.0 was taking shape and World of Warcraft was conquering the globe, Curse rode multiple waves of gaming evolution: from the mod revolution to the wiki wars, from voice communication innovation to the streaming explosion. Each pivot wasn't just reactive—it was prescient, often arriving just ahead of where the market was heading.

↑ Back to Top

II. The Origin Story: Addiction to Ambition

The legend begins with obsession. As a teenager, Thieblot was a fan of World of Warcraft, a popular massively multiplayer online roleplaying game. But "fan" understates it dramatically. This was total immersion—the kind that destroys sleep schedules, social lives, and eventually, family relationships.

The breaking point came when Thieblot's parents reached theirs. Playing World of Warcraft for 16 hours daily wasn't sustainable under their roof. The ultimatum was clear: the game or the home. Thieblot chose the game. Yet this apparent disaster became the crucible for entrepreneurship. After leaving school, Thieblot began to turn his passion into a business, launching CurseBeta in 2006, offering up add-ons and modifications.

The early days were quintessentially startup—operating on passion more than capital. As the company began to increase in traffic and revenue, Thieblot hired his brother as the first Curse employee, and after relocating to Germany, hired a third developer to aid in the site development. This wasn't Silicon Valley polish; this was European scrappiness meeting American opportunity.

The initial insight was elegant in its simplicity: World of Warcraft's modding community was massive but fragmented. Players needed add-ons to customize their gameplay experience, but finding, downloading, and updating them was a nightmare of forum posts, broken links, and version conflicts. Thieblot didn't just understand this problem intellectually—he'd lived it for thousands of hours.

By late 2006, the site exponentially increased in traffic and popularity. What started as one obsessed gamer's solution to his own problem was resonating with millions facing the same frustrations. The WoW community wasn't just large; it was passionate, technical, and willing to pay for solutions that enhanced their gameplay experience.

↑ Back to Top

III. Finding Product-Market Fit: The Mod Repository Era (2006-2009)

Building a mod database was one thing. Making it indispensable was another. The breakthrough came in 2007 with a product that would define Curse's early success: Thieblot began development on the Curse Client, a comprehensive solution to add-on management and distribution.

The Curse Client solved a problem every serious WoW player knew intimately. Before its arrival, updating mods meant manually checking dozens of websites, downloading files, unzipping them to the right folders, and praying nothing broke. The Client automated this entire process—one click to update all your add-ons. As the client matured into beta status, its feature set became more comprehensive, supporting connections to various Curse modification databases.

The product-market fit was immediate and obvious. Thieblot says that Curse currently has over 34,000 paid subscriptions since launching the premium option eight months ago. Players weren't just using the free version—they were paying for premium features like the ability to update all of your plugins simultaneously and allowing users to save their UI setup to the server, allowing them to restore it should they start playing from a friend's house or during a break at school. Gamers can spend a very long time mapping out exactly where each of their UI elements and shortcuts appear on the screen.

But the real strategic insight came in 2008. With over 2/3rds of his traffic coming from the United States, Thieblot moved Curse to San Francisco and began to expand Curse into community sites, news, forums, databases, and original content. This wasn't just following the traffic—it was recognizing that to build a lasting business in gaming, you needed to be where the industry's heart was beating.

The move coincided with Curse's first major funding rounds. A $5 million Series A round in 2007 led by AGF Private Equity, followed by a $6 million Series B round in early 2009 with participation from Ventech Capital, AGF Private Equity, and SoftTech VC (Jeff Clavier), bringing Curse's total funding to $11 million.

The timing was fortuitous. Thieblot has managed to turn his addiction into a thriving company called Curse that generated over $3 million in revenue this year. By 2009, Curse wasn't just surviving—it was thriving with 7.4 million uniques monthly and a clear path to profitability.

↑ Back to Top

IV. Scaling the Gaming Media Empire (2010-2013)

The period from 2010 to 2013 marked Curse's transformation from a successful tools company into a gaming media empire. The strategy was aggressive: build and buy everything that hardcore gamers needed.

Recognition came quickly. In 2010, Curse became a Microsoft Bizspark One company, joining an elite program for high-growth startups. In 2011, Inc. 500 ranked Curse Inc. as the 405th fastest growing company in the United States, and the San Francisco Business Times ranked it 22nd in their list of the "Top 100 Fastest Growing Companies in the San Francisco Bay Area".

In April 2012, Ernst & Young named Thieblot as a semifinalist in their "Ernst & Young Entrepreneur of the Year" program for Northern California, recognizing his entrepreneurial efforts and successes with Curse. By June 2012, Curse's monthly worldwide traffic was reported by Quantcast as being in excess of 21 million unique visitors.

The acquisition strategy was methodical. Curse owns blogs, wikis, community sites, and download hubs for a number of other popular games. Curse has built out about half of the communities itself, and it also actively acquires leading websites related to gaming. For example, Curse acquired DiabloFans.com, which is currently the second ranked Google match for "Diablo 3". Diablo 3 wasn't actually out yet, but this would be prime real estate as soon as it launched. This forward thinking is a big part of the company's strategy: Thieblot says that Curse tries to stay ahead of the curve, strategically trying to figure out which games are going to be hits and then positioning themselves accordingly.

December 2012 brought one of Curse's most ambitious moves: Curse officially launched the Gamepedia wiki farm. This wasn't just another acquisition—it was a direct challenge to Wikia in the gaming space. The platform would grow to host as of April 2019, 1,293,790 contributors, 6,224,464 articles and 2,195 wikis. High-profile wikis such as The Official Witcher Wiki, the Official Minecraft Wiki, The Official ARK: Survival Evolved Wiki, Dota 2 Wiki, Leaguepedia, COD Wiki, and Wowpedia had hundreds of thousands of edits across thousands of accounts. These wikis were also available in several languages.

Then came a surprise move that would define Curse's next chapter. On June 26, 2013, Curse announced in a press conference that it would relocate their chief headquarters to Huntsville, Alabama, leaving their sales office in San Francisco. The decision shocked Silicon Valley observers, but the logic was sound. "I did an analysis for all of those places under consideration for corporate tax rates, the availability of talent, the cost of living and employee taxes," said Curse VP of Finance Brandon Byrne. "As it turned out, Huntsville was very competitive."

The move to Huntsville represented more than cost savings—it was a bet on sustainable growth over Silicon Valley hype. Curse has grown at a rate of more than 50 percent a year for the past four years. Curse has 100 employees worldwide, including offices in Germany, France, United Kingdom, Australia, Ireland and Canada. The Alabama location offered something Silicon Valley couldn't: the ability to build without burning through capital on astronomical rents and salaries.

↑ Back to Top

V. The Curse Voice Pivot: Gaming's Communication Revolution (2014-2015)

Sometimes the biggest opportunities hide in the most overlooked problems. By 2014, Thieblot recognized that voice communication in games had stagnated. TeamSpeak, Ventrilo, and Mumble had barely evolved in years. They were functional but clunky, requiring players to alt-tab out of games, share IP addresses, and manage server passwords. For a generation raised on seamless mobile experiences, this felt archaic.

In 2014, Thieblot announced the beta version of Curse Voice, a desktop Voice over IP (VOIP) application designed for games such as League of Legends, integrating voice chat and auto-match making with several titles. The timing seemed perfect—or so they thought.

The launch was nothing short of spectacular. Curse launched an open beta of Curse Voice for Windows in May 2014 and had a million users in its first week. If it continues to grow at its current rate, Curse anticipates the service will hit 10 million active users by year's end.

The innovation wasn't just in the technology—it was in the approach. The software can automatically detect users of the same group and prompt them to connect to a voice chat session. Other voice clients require users to share specific account credentials with other players. That can be cumbersome on a match-by-match basis, and it opens users up to security risks by making their Internet protocol addresses available. Curse Voice makes connecting more convenient and risk free. The program automatically identifies teammates who are using the software and enables them to enter a voice session with the click of a button without ever revealing personally identifiable information.

By May 2015, the metrics were impressive: Curse Voice has now attracted 1.1 million monthly active users (of 3.5 million total sign-ups), who have formed 16 million "friendships" and sent 92 million messages. In June 2015, Curse Voice expanded its offerings by releasing apps for Mac, Android and iOS, giving users more ways to use the service while on the go or at home.

The product was gaining serious traction with game developers too. The Robocraft integration showcased the potential: After integrating Curse Voice, Robocraft immediately saw a 258% increase in active Robocraft players. This wasn't just about voice chat—it was about reducing friction in multiplayer gaming and keeping players engaged longer.

Then came the validation that changed everything. On July 7, SEC reports revealed that Riot Games had invested $30 million in Curse. This amounts to more than Curse's initial round of $12 million funding and series B $16 million funding. For context, Riot Games wasn't just any investor—they were the creators of League of Legends, the world's most played PC game.

"Riot really believes in what we're doing and where we're going," said Thieblot. "This investment boosts our efforts to continue creating what I believe to be the best communication tool in gaming. In the last year, we've seen the size of Curse Voice grow from nothing to over 1 million active players."

The Riot investment was particularly meaningful because of the companies' complicated history. Just a year earlier, Riot had banned Curse Voice from League of Legends, concerned that certain features gave unfair advantages. Curse quickly removed the offending features and worked to align with Riot's vision. The investment signaled not just forgiveness but embrace—Riot saw Curse Voice as potentially integral to League of Legends' future.

By the end of 2015, Curse had built something remarkable: a communication platform designed specifically for gamers, backed by the industry's biggest player, with millions of users and a clear path to monetization. The company had raised approximately $52-58 million total and was generating significant revenue from both its media properties and software subscriptions.

Yet unknown to Curse, a competitor was rising. Discord launched in May 2015, just as Curse Voice was hitting its stride. While Curse had first-mover advantage, enterprise partnerships, and Riot's backing, Discord had something else: a relentless focus on user experience and a willingness to burn venture capital to achieve dominance.

↑ Back to Top

VI. The Amazon/Twitch Acquisition (2016)

The online retailing giant said Tuesday that it had bought Curse, an online media and gaming community. Financial details of the acquisition were not disclosed. The announcement came as Curse was at a crossroads. Despite the success of Curse Voice and strong revenue from its media properties, the competitive landscape was shifting rapidly.

Founded in 2006, Bay Area-based Curse had raised nearly $60 million to date for its platform that gives users PC gaming-related information and add-ons like videos, guides, forums, communication apps, streaming tools, mod managers, and more. Curse operates 65 gaming sites including ones focused on news, guides, tools, and online communities that attract more than 30 million people monthly.

The strategic logic for Twitch was clear. Amazon acquired Twitch, used by gaming enthusiasts to watch others play online games such as Minecraft and Call of Duty, for around $1 billion in 2014. Twitch gained traction before the rise of live video streaming services like Facebook Live and Twitter's Periscope. Twitch now has 100 million users.

"We've long been fans of Curse, which is an innovator in the games industry with a strong culture built around its offerings—from Curse Voice and Curse Client to Gamepedia," Emmett Shear, Twitch's CEO, said in a statement. "I'm really excited to see how we can bring Curse services into the Twitch network, and provide an unparalleled experience to both Curse and Twitch users," commented Hubert Thieblot, founder and CEO of Curse.

What made the acquisition particularly interesting was what happened next. In 2016, Curse Voice was rebranded and redesigned as Curse. It brought modding support for more games such as WoW. After being acquired by Twitch, the Curse app received an update adding the ability to "sync" Curse accounts with Twitch accounts. The Curse app added screen sharing and video calling in 2016. On March 16, 2017, the Curse app was rebranded as the Twitch Desktop App and received a redesign.

This rebrand revealed Amazon's true ambition: to challenge Discord directly. VentureBeat stated that this was an attempt to compete with Discord, "the dominant social platform in the gaming space." But the integration was troubled from the start. The Twitch Desktop App felt like two products awkwardly merged rather than one cohesive vision.

The problems became apparent quickly. The Twitch Desktop App removed VOIP features in February 2019. After less than two years, Amazon essentially admitted defeat in the voice communication space. Discord had won that battle decisively.

↑ Back to Top

VII. The Great Unbundling (2017-2020)

If the Twitch acquisition was meant to create a gaming super-platform, what followed was a methodical dismantling of the empire Thieblot had built. Each piece found a new home, but the vision of an integrated gaming ecosystem died.

The first major split came in December 2018. Fandom, Inc. had reached an agreement to acquire the spin-off of Curse called Curse Media from Twitch Interactive for an undisclosed amount. This included Gamepedia, Curse Network, D&D Beyond, Muthead, and Futhead. Fandom says it expects to close it by early 2019.

The Gamepedia sale was particularly symbolic. Gamepedia is one of the jewels in its cap and, according to SimilarWeb, earns an average of 133 million monthly visits. What Curse had built as a Wikia competitor was now being absorbed by Wikia's successor, creating the undisputed champion of gaming wikis.

Then came the final piece. Overwolf, the in-game app-development toolkit and marketplace, has acquired Twitch's CurseForge assets. In June 2020, Overwolf acquired CurseForge, for an undisclosed sum, from Twitch. As a result of the acquisition, the CurseForge mod manager will move from being a Twitch client and become a standalone desktop app included in Overwolf's suite of app offerings.

The Overwolf acquisition was revealing. As the company makes its pitch to current CurseForge users — hoping that the mod developers will stick with the marketplace, they're offering to increase by 50% the revenue those developers will make. This suggested that under Twitch's ownership, CurseForge hadn't been optimized for creator monetization—a surprising oversight given Amazon's usual focus on marketplace dynamics.

By 2020, the Curse empire had been fully dismantled: - Curse Voice had been absorbed into Twitch and then abandoned - Gamepedia belonged to Fandom - CurseForge was owned by Overwolf - The various community sites were scattered across multiple owners

He now works at Twitch as Vice President of Emerging market/Mobile first product. Thieblot remained at Twitch through January 2018, but the company he built no longer existed as a coherent entity.

↑ Back to Top

VIII. Porter's Five Forces Analysis

Analyzing Curse through Porter's framework reveals why the company was both valuable and vulnerable:

Supplier Power (Medium-High): Game developers held significant leverage. Curse needed official approval for its wikis and tools, and games could change their APIs or terms of service at will. The Riot Games episode with Curse Voice—first banning it, then investing $30 million—exemplified this dynamic.

Buyer Power (Low): Individual gamers had minimal bargaining power. While switching costs between platforms were low, the network effects of established communities created stickiness. Once a guild used Curse Voice or a wiki became authoritative, migration was painful.

Threat of Substitutes (High): Every Curse product faced substitutes. CurseForge competed with NexusMods and Steam Workshop. Curse Voice faced TeamSpeak, Ventrilo, Mumble, and ultimately Discord. Gamepedia competed with Wikia/Fandom. This constant substitution threat limited pricing power.

New Entrants (Mixed): Barriers varied by product. Creating a basic gaming website or forum had low barriers, but building voice technology or achieving wiki critical mass required significant investment. Discord's entry and rapid dominance showed that with enough capital, barriers could be overcome.

Competitive Rivalry (Intense): Gaming communities are passionate and fragmented. Every popular game spawned multiple competing websites, tools, and communities. Consolidation through acquisition was common because organic growth was difficult in such a crowded space.

↑ Back to Top

IX. Hamilton's Seven Powers Framework

Applying Hamilton Helmer's Seven Powers reveals Curse's strategic positioning:

Scale Economies (Strong): With 30+ million unique visitors by 2013, Curse had significant scale advantages in advertising sales and infrastructure costs. The company's ability to monetize traffic across multiple properties created operational leverage.

Network Effects (Moderate): Curse wikis and communities exhibited network effects—more contributors meant better content, attracting more users. However, these effects were localized to individual games rather than platform-wide, limiting their defensive power.

Counter-positioning (Failed): Curse Voice attempted counter-positioning against traditional VOIP services with gaming-specific features and auto-connection. Discord later succeeded with this exact strategy, suggesting Curse's execution rather than strategy was flawed.

Switching Costs (Weak): Users could easily switch between mod managers, wikis, or voice clients. The main switching cost was the inconvenience of reinstalling mods or losing social connections, but these proved insufficient against superior alternatives.

Branding (Limited): Curse had strong brand recognition within specific communities (WoW modding, Minecraft) but never achieved broader gaming mindshare. The multiple rebrandings under Twitch diluted whatever brand equity existed.

Cornered Resource (Temporary): Exclusive partnerships with game developers for official wikis and the Riot Games relationship represented cornered resources. However, these proved temporary as developers could switch partners or bring capabilities in-house.

Process Power (Moderate): Curse developed a repeatable playbook for acquiring and integrating gaming websites. The ability to identify undervalued communities, acquire them cheaply, and monetize them through the broader network was a genuine capability, though not ultimately defensible.

↑ Back to Top

X. Key Inflection Points & What-Ifs

Several critical decisions shaped Curse's trajectory:

2008 - The Geographic Move: Moving from Europe to San Francisco was expensive but necessary. Staying in Europe would have limited access to venture capital and talent. However, the 2013 move to Huntsville suggests the SF location wasn't essential for operations.

2012 - Gamepedia Launch: Creating a wiki farm put Curse in direct competition with Wikia. While Gamepedia succeeded, it fragmented resources. Partnering with Wikia instead might have allowed focus on higher-margin products.

2014 - Curse Voice Timing: Launching just one year before Discord was catastrophic timing. Earlier launch might have established market dominance; later launch might have revealed Discord's momentum and suggested pivoting resources elsewhere.

2015 - Riot's Investment: The $30 million from Riot validated Curse Voice but may have also constrained it. Riot's specific needs for League of Legends might have prevented Curse from building the general-purpose platform Discord became.

2016 - Selling to Amazon: Selling to Amazon/Twitch seemed logical but proved problematic. An independent Curse might have pivoted more aggressively to compete with Discord or found other growth avenues.

The biggest what-if: What if Curse had recognized Discord's threat earlier and pivoted fully to voice in 2015? With Riot's $30 million, first-mover advantage, and existing user base, Curse could have potentially won the voice communication market worth billions today.

↑ Back to Top

XI. Playbook: Lessons for Founders

Curse's journey offers valuable lessons:

Start with Genuine Passion: Thieblot's World of Warcraft obsession wasn't manufactured. Deep, authentic understanding of user problems enabled rapid product-market fit.

Solve Your Own Problem First: The Curse Client succeeded because Thieblot had personally experienced the pain of managing WoW add-ons thousands of times.

Geographic Arbitrage Works: The Huntsville move proved that gaming companies don't need Silicon Valley. Lower costs enabled profitability and sustainable growth.

Horizontal Expansion Has Limits: Curse's empire-building across wikis, forums, tools, and voice spread resources thin. Deeper focus on fewer products might have created more defensible positions.

Timing Matters More Than Features: Curse Voice was arguably superior to early Discord but launched at the wrong moment. In winner-take-all markets, being early or late by even months can be fatal.

Corporate Acquisition Often Means Dismemberment: Large acquirers rarely maintain entrepreneurial visions. Amazon/Twitch bought Curse for parts, not the whole.

Creator Monetization Is Undervalued: Overwolf immediately increased creator payments 50%, suggesting Curse/Twitch undermonetized. Aligning platform success with creator income creates powerful incentives.

↑ Back to Top

XII. Grading & Final Analysis

Was the Amazon acquisition successful? It depends on perspective.

For Thieblot and investors, absolutely. While the exact price wasn't disclosed, selling a company after raising ~$58 million suggests a successful exit. Thieblot transformed a gaming addiction into generational wealth.

For Amazon/Twitch, the verdict is mixed. CurseForge and Gamepedia were valuable assets that enhanced Twitch's gaming ecosystem. The attempt to challenge Discord failed completely, but the acquisition price was likely modest relative to Amazon's resources.

For the gaming community, the unbundling was probably optimal. Fandom has the resources to maintain Gamepedia properly. Overwolf is incentivized to innovate with CurseForge. Specialized owners for specialized products makes sense.

The real tragedy is Curse Voice. The product had genuine innovation, strong partnerships, and early traction. In an alternate timeline where Discord doesn't exist or Curse executes better, voice communication in gaming might look very different today.

Thieblot's post-acquisition role as VP of Emerging Markets at Twitch suggests ongoing influence, but his entrepreneurial journey with Curse ended with the acquisition. The empire he built from a World of Warcraft obsession has been parceled out, but its components continue powering gaming experiences for millions.

Could this playbook work again? Unlikely in the same form. The gaming landscape of 2006-2016 was uniquely fragmented, allowing roll-up strategies. Today's gaming ecosystem is more consolidated, with platform owners like Discord, Steam, and Epic controlling distribution. The next gaming empire won't be built through acquisition and aggregation but through platform disruption or new modalities (VR, AI, blockchain gaming).

Final Grade: B+

Curse achieved remarkable success—from teenage addiction to successful exit, from zero to 30 million users, from Huntsville to Amazon. It pioneered gaming wikis, modernized mod distribution, and nearly revolutionized voice chat. The company generated real value for millions of gamers and meaningful returns for investors.

Yet it fell short of its ultimate potential. Curse Voice could have been Discord. Gamepedia could have remained independent. The empire could have stayed intact. These aren't failures exactly, but they represent unrealized possibilities.

In the end, Curse's story is quintessentially entrepreneurial: a passionate founder solving personal problems, riding market waves, making bold bets, and achieving a successful if imperfect outcome. Thieblot turned an addiction into an empire, even if that empire eventually dissolved into the broader gaming ecosystem. For a company born from being kicked out for playing too much World of Warcraft, that's a legendary achievement.