The Untold History of the Offerwall

In October 2009, the offerwall industry nearly ceased to exist. It was the second time in three years that the model had been brought down by a scandal, and neither one was about the thing people assumed.

Most people working in rewarded advertising today have no particular reason to know any of this. The category is stable now — verified actions, publisher vetting, disclosure requirements, fraud controls. Those things feel like obvious engineering choices rather than the settlement of two separate crises. But nearly every convention in a modern offerwall is a scar, and the story starts earlier than almost anyone assumes: not with a game, but with a website in 2004 offering strangers a free iPod.

This is that history, and what it explains about how offerwalls work today.

Before the offerwall: the free iPod era

The offerwall is older than the smartphone, older than Facebook applications, and older than the phrase “offerwall.” Its true consumer debut was a website that most people assumed was a scam and which turned out not to be.

In the summer of 2004, a Washington DC company called Gratis Internet — run by Peter Martin and Rob Jewell — launched FreeiPods.com. The proposition was exactly what the domain said. Complete one advertiser offer from a list of about ten, get five friends to do the same, and Gratis would ship you an iPod. The offers came from mainstream advertisers: Blockbuster, Columbia House, Citigroup credit cards. Most were free trials that could be cancelled once the trial ended.

The internet’s reaction was near-universal disbelief, and the disbelief was wrong. Student newspapers and consumer reporters investigated it repeatedly through late 2004 and kept reaching the same conclusion: the iPods were real and they arrived. Gratis was shipping roughly 500 a week, and by that autumn had sent out something on the order of $2.5 million in merchandise.

The economics are worth understanding, because they are the offerwall’s economics. Gratis earned roughly $40 per completed offer. An iPod cost around $300. A single user completing a single offer was deeply unprofitable — but that user brought five more, each of whom completed an offer and then tried to recruit five of their own. The referral tree, not the individual completion, is what paid for the hardware. It worked spectacularly: within about two months, monthly revenue went from around $300,000 to $3 million.

Success bred a family of sites — FreeFlatScreens.com, FreeDesktopPC.com, and others on the same template — and in August 2005 Gratis consolidated them all under the FreePay brand.

Every structural element of a modern offerwall is already visible here. A user who won’t pay cash pays with an action instead. An advertiser pays for a verified completed lead rather than an impression. An intermediary sits between them, aggregates demand, and takes the spread. And the reward is tiered against escalating effort — one offer gets you into the system, several get you the prize — which is the direct ancestor of the multi-reward structures used today.

The first trust failure — and it wasn’t about the offers

The free-stuff model didn’t collapse because it failed to deliver. It collapsed over something users never saw.

From 2000 through 2004, Gratis’s sites carried an explicit privacy promise: the company would never give out, sell, or lend user information to anyone. In March 2006, New York Attorney General Eliot Spitzer announced the outcome of an investigation finding that Martin and Jewell had sold access to as many as seven million confidential user records to email marketers, in direct contradiction of that promise. Investigators described it as the largest deliberate breach of a privacy policy then discovered by US law enforcement. Datran Media, one of the recipients, paid $1.1 million in penalties, disgorgement and costs under an Assurance of Discontinuance.

This matters to the story for a reason that becomes obvious three years later. The reward model’s vulnerability was never really the merchandise. It was that a business built on collecting user actions is also a business collecting user data and user trust, and both can be quietly monetized a second time without anyone noticing until a regulator does.

The move into games

What changed in the late 2000s was the arrival of a better host environment. Facebook and MySpace applications gave millions of people games with virtual economies, and those economies created a problem worth solving: players wanted currency, most wouldn’t pay for it, and payment processing for small transactions was clumsy and expensive.

Offer-based monetization answered that neatly, and it no longer needed a referral tree to work. The reward was virtual currency rather than hardware, so it cost the developer almost nothing to mint. A player who wouldn’t enter a credit card would complete a survey, sign up for a trial, or subscribe to something in exchange for enough currency to keep playing. Companies formed quickly to intermediate it — Offerpal Media, Super Rewards, Gambit and others — and by 2009 offer revenue was a substantial share of income for the largest social game developers.

RevU entered in 2005, mid-way through the free-stuff era, monetizing online games and loyalty sites. That timing matters to the rest of this story mainly because it means RevU was already operating before either reckoning arrived, and is still operating after both.

Scamville

The problem was the offers themselves.

A meaningful portion of the inventory flowing through offerwalls in 2009 was not honest. Users clicked through to what looked like a simple quiz or a free trial and ended up enrolled in mobile subscriptions billed to their phone, or in negative-option programs that were far easier to enter than to leave. The reward was real. The cost was concealed. Because the intermediary was paid per completed lead, the incentives pointed the wrong way for everyone except the advertiser running the deceptive offer.

The confrontation became public at the Virtual Goods Summit in October 2009, where TechCrunch editor Michael Arrington challenged Offerpal CEO Anu Shukla directly about the legitimacy of her company’s offers. She denied the characterization forcefully. Arrington then published “Scamville: The Social Gaming Ecosystem Of Hell” on October 31, arguing that users were being tricked into lead-generation scams and that the social gaming economy was structurally dependent on it.

What followed was fast and industry-wide.

• Within a week, Zynga announced it would remove all in-game offers. Facebook had already blocked the launch of one of its titles until controversial offers were removed.

• Offerpal replaced its CEO. Shukla departed in early November and was succeeded by George Garrick, who publicly conceded that the company had been guilty of distributing offers of questionable integrity from some of its advertisers — a remarkable admission from a sitting chief executive about the business he had just inherited.

• A proposed class action named Facebook, MySpace, Zynga and others.

• MySpace, RockYou and other platforms rewrote their policies on what offers could appear in applications.

• Offerpal published a set of offer standards intended to lock out the worst inventory.

The category did not die. But it emerged with a permanent reputational scar and, more usefully, with a set of rules.

What the crisis actually fixed

This is the part that gets lost. The 2009 collapse wasn’t just a public relations event — it produced the structural conventions that define the format today.

Advertiser vetting became a network responsibility. Before Scamville, offerwalls largely passed through whatever demand their affiliate sources supplied. After, the network was expected to know what it was serving and to be accountable for it. Publisher-facing offer standards date from this period.

Disclosure moved to the front. The core deception had been about what completing an offer would cost. Showing requirements clearly before a user starts — what they must do, what they’ll be charged, what they receive — became a baseline expectation rather than a courtesy.

The subscription trap became radioactive. Negative-option and mobile-billing offers were the specific mechanism behind most of the harm, and their removal from mainstream offerwalls dates to this moment.

Verified actions displaced raw leads. Paying for a submitted form invites fraud and low-quality demand alike. Paying only when a verified downstream action occurs — an install with real engagement, a completed signup, a subscription that reaches billing — aligns everyone. This idea long predated 2009 in performance marketing, but the offerwall category’s wholesale move toward it accelerated sharply afterward.

User data stopped being a second product. This one is inherited from the Gratis settlement rather than from Scamville, and it’s the reason mainstream offerwalls today pass a completion signal rather than a marketable identity. The lesson of 2006 was that a rewarded ad business sits on a large pool of user records, and that quietly selling them is both very tempting and career-ending.

If you have ever wondered why offerwall networks talk so much about validation and publisher vetting, this is the answer. Those aren’t marketing themes. They’re the terms on which the industry was allowed to continue.

The mobile transition, and a second shock

Offerpal rebranded as Tapjoy in 2010 and pivoted to mobile, where the app stores were producing exactly the conditions that had made social gaming lucrative: virtual economies, price-sensitive users, and developers desperate for installs.

That worked until 2011, when Apple moved against incentivized installs influencing App Store chart rankings. Paying users to install an app had become an efficient way to buy your way up the charts, and Apple’s intervention removed the single largest use case for offerwall inventory more or less overnight.

The category adapted again, and the adaptation is visible in how offerwalls are priced now. If you cannot sell installs-for-ranking, you have to sell something an advertiser values on its own merits — engagement, subscriptions, purchases, retention milestones. The industry’s move toward cost-per-engagement and cost-per-action pricing is a direct descendant of Apple’s 2011 decision. Two of the format’s defining characteristics, in other words, were imposed from outside during crises rather than designed.

Consolidation

The last decade of offerwall history is mostly an acquisition table.

Tapjoy — the company that began as Offerpal — was acquired by ironSource in January 2022, and ironSource was itself acquired by Unity later that year. The Tapjoy offerwall still operates inside Unity’s ecosystem. Fyber, another major offerwall operator, was acquired by Digital Turbine. Adscend Media, founded in 2009 at the height of the controversy, is now part of the Edge226 group. Newer entrants arrived with different mechanics: adjoe, founded in Hamburg in 2018 and backed by applike group and Bertelsmann, built its business on rewarded playtime rather than discrete completions.

The pattern is that offerwalls have largely become features of larger platforms rather than standalone businesses. That has real consequences for publishers, and they aren’t all bad — scale brings demand. But an offerwall inside a mediation platform competes for roadmap attention with every other product in the suite, and its strategic priorities are set by the parent.

RevU is the outlier here, and is the reason it can accurately describe itself as the oldest continuously-operating offerwall in the gaming ecosystem. Not the oldest brand — several of those names are older as trademarks. The oldest one that has run continuously, independently, under its own strategy, without being absorbed into a larger stack.

Why any of this matters to a decision you’re making today

Three practical things follow from the history.

Institutional memory is a real asset in this category. The failure modes are not new. Deceptive offers, subscription traps, incentive abuse, ranking manipulation — each has been seen, litigated, and addressed before. A network that lived through those cycles has calibrated policies rather than inherited ones.

“Incentivized” still carries a stigma the format has largely outgrown. When an advertiser dismisses rewarded traffic as low quality, they are frequently reacting to a reputation formed in 2009 about practices that mainstream networks abandoned that same year. The actual measurement questions around rewarded traffic are quite different from the ones that reputation implies.

The old temptations recur. Every mechanism that caused the 2009 crisis still works in the narrow sense of extracting more short-term revenue. Concealed requirements, engineered reward friction, and misleading offer titles all raise near-term numbers. The industry’s conventions exist to stop that, and conventions decay when nobody remembers what they were for. We’ve written separately on how these patterns resurface in modern interfaces.

FAQs

When were offerwalls invented?

The mechanic reached consumers in 2004 with Gratis Internet’s FreeiPods.com, where users completed advertiser offers and recruited friends in exchange for physical merchandise. It took its recognizable modern form — offers exchanged for virtual currency inside a game — on Facebook and MySpace applications around 2007–2009. RevU began monetizing games and loyalty sites in 2005, between the two.

Was FreeiPods.com a scam?

No, which surprised nearly everyone at the time. Journalists and student newspapers investigated it repeatedly in 2004 and found the iPods were genuinely delivered — Gratis Internet was shipping around 500 a week. The company’s actual legal trouble came later and from a different direction: a March 2006 New York Attorney General settlement over selling user records to email marketers in violation of its own privacy policy.

How did the free iPod sites make money?

Gratis earned roughly $40 for each completed advertiser offer, against an iPod costing about $300. The referral requirement is what closed the gap — each recipient brought five more people who each completed an offer and then recruited their own, so the cost of the hardware amortized across a growing tree rather than a single user. Monthly revenue went from roughly $300,000 to $3 million within about two months of launch.

What was the Scamville controversy?

In October 2009, TechCrunch published “Scamville,” arguing that social gaming revenue depended heavily on deceptive lead-generation offers that enrolled users in concealed subscriptions. The fallout included Zynga removing all in-game offers, a CEO change at Offerpal Media whose successor publicly admitted the company had distributed offers of questionable integrity, a threatened class action, and industry-wide policy reform.

Why did offerwalls move from paying per install to paying per action?

Two reasons. Apple’s 2011 move against incentivized installs influencing App Store rankings eliminated the biggest buyer of install volume. And paying only for verified downstream actions produces better advertiser outcomes while removing most of the economic incentive for fraud, since faking a deep action costs more than the payout is worth.

Is Tapjoy still in business?

The Tapjoy offerwall still operates, now within Unity’s ecosystem. Tapjoy — originally Offerpal Media — was acquired by ironSource in January 2022, and Unity acquired ironSource later the same year.

Which is the oldest offerwall still operating?

RevU, which has run continuously since 2005 and remains independently operated. Several older brand names survive, but as products inside larger platforms following acquisitions.

The takeaway

The offerwall is one of the few advertising formats that was publicly accused of being fundamentally dishonest, largely deserved the accusation, and then reformed rather than disappearing. It happened twice — once over user data in 2006, once over deceptive offers in 2009 — and the verification, vetting, and disclosure practices that define the format now were not clever product decisions. They were the price of survival, paid in instalments across 2006, 2009 and 2011.

The through-line is worth naming. What nearly killed the model was never the core exchange, which users understood perfectly well and largely liked. It was what got attached to it when nobody was looking — concealed subscription terms, and user records sold out the back. The free iPods arrived. That was never the problem.

Two decades in, RevU is still running the same business it started in 2005 — which is, in a category this turbulent, the least common outcome of all. More about RevU, or talk to our team.