
Loopt, based in Silicon-Valley and backed by leading venture capital firms Sequoia Capital and New Enterprise Associates, has created an interoperable social-mapping service that allows individuals to use their location to discover the real world around them – enabling them to find and enjoy the people, places, and events that mean the most right here and now using their mobile phones. With Loopt, individuals will always know who’s around, what to do, or where to go
Explore the risks and possibilities with a prompt for ChatGPT, Claude, or your agent.
Loopt let people find nearby friends through their phones. Founded in 2005 by Sam Altman, Nick Sivo, and Alok Deshpande, it joined Y Combinator’s first summer batch and became an early example of mobile social mapping. Its product grew from friend locations into local discovery, check-ins, and merchant rewards. 1 2
Its trajectory exposes a difficult gap between distributing a location feature and building a business around it. Carrier deals and an iPhone launch created access, but the public evidence does not establish durable engagement or profitable local commerce. That distinction matters more than a simple verdict that Loopt was too early.
Green Dot announced its purchase in March 2012 for $43.4 million, including $9.8 million reserved for employee retention. The consumer products were to close while the team moved into mobile financial services. This was an acquisition with a consumer-product wind-down, rather than a bankruptcy. 3 4
Image 1 / 2
The founding problem came from Stanford students calling friends to ask where they were. A map could answer that question before the call. Contemporary reporting identifies Sivo and Deshpande alongside Altman; it also describes the company’s progression from a student project into carrier-backed distribution. 2
Altman’s later account gives the decision more texture. In the Masters of Scale interview published by YC, he said, “I fell into it accidentally.” He had enjoyed research in Stanford’s computer science department and worked on the project after classes. He had accepted a Goldman Sachs internship, then chose the project and YC instead. He said, “It would not have been a startup if it were not for Y Combinator.” 5
That origin helps explain both the ambition and the risk. A useful student project can attract investment before its founders know whether the use case supports a large company. Altman later acknowledged that he had not given that question enough deliberate attention. His account supports a lesson about reassessing the business as it grows; it does not establish which individual decision caused the exit. 5
The original experience showed friends on a map and notified users when people were nearby. Permission controls let users hide their locations. The appeal was practical: know who is close enough to meet without asking everyone separately. Loopt Mix later added discovery of nearby people with shared interests. 2
Background location did not arrive automatically with the 2008 iPhone app. Loopt initially faced the iPhone’s restriction on third-party background execution. It announced a network-based workaround in 2009. Apple introduced multitasking services in 2010, and Loopt’s updated app used iOS 4. These were different implementations at different dates. 2 11 9
The product also moved beyond passive friend finding. Pulse combined places and events; its BlackBerry update drew recommendations from Zagat, Citysearch, Tasting Table, Metromix, Zvents, and SonicLiving. Star used check-ins to unlock merchant rewards, including Gap offers. Those experiments directly contradict an account in which Loopt never tried active sharing or commerce. 12 13
Loopt also integrated Facebook Places, displaying Facebook friends’ check-ins and supporting shared updates across both services. Facebook was therefore both a competitive threat and a source of distribution and data. 14
The initial carrier route reached people using supported phones, including Boost Mobile customers. The later product served smartphone users who wanted friend discovery, local recommendations, or rewards. These are overlapping jobs, but they require different reasons to return. A nearby friend can trigger a meeting; a discount can trigger a purchase. Combining them does not guarantee either outcome.
A credible historical revenue estimate for Loopt’s specific addressable market is unavailable in the reviewed evidence. Mobile adoption created a larger potential audience, but device counts cannot establish willingness to pay for friend coordination. The useful question was how frequently a reachable group of friends would use the product together.
Loopt faced Google Latitude, Brightkite, and Whrrl by 2009. Later products competed through check-ins and rewards. 2 The structural challenge was the social group: the map becomes useful when relevant friends participate, rather than when an isolated user installs it.
Facebook Places added location check-ins in August 2010. 15 Loopt responded through integration. That response makes a categorical claim that Facebook instantly killed Loopt too strong. An established social network reduced the cost of finding friends, while a separate app still needed a reason to become their regular destination.
Loopt tested several ways to earn from location. The 2009 network service charged a monthly fee. In 2010, Altman described banner advertising and growing merchant interest in loyalty rewards. Star supplied a concrete merchant experiment rather than an untested advertising idea. 2 8 13
The missing evidence is the economics: attributable revenue, repeat purchasing, customer acquisition costs, and gross margins. Merchant participation proves that a program launched. It does not prove that rewards generated enough incremental spending to sustain Loopt. The reviewed sources do not supply that calculation.
In April 2010, Altman told The Next Web that Loopt had more than three million users and was growing above 15% monthly. Those were management claims, with no retention series attached. 8
VentureBeat reported 500 daily users in 2012; Altman disputed it. That disputed figure cannot establish Loopt’s active audience. 16
Nor does the number of employees joining Green Dot establish a prior layoff rate. YC’s announcement described 30 joining employees; an earlier database headcount is not a comparable payroll record. 4 Registered accounts, claimed growth, and acquisition staffing measure different things.
Loopt solved real access problems through carriers and later smartphone support. The 2008 iPhone platform still constrained its central behavior. That is evidence of platform dependence, but the product remained active after the 2010 improvement. Timing alone cannot explain the eventual sale.
The stronger interpretation is that broad availability did not establish a durable, monetizable group habit. This remains an inference: retention and unit economics are missing. It would weaken if reliable records showed strong repeated group usage and profitable commerce before acquisition.
The 2008 iPhone launch produced complaints about confusing SMS invitations and opt-outs; Loopt subsequently changed the invitation flow. 6 The episode matters because a product handling locations needs clear consent. The available evidence does not measure its lasting effect on retention. It also does not support treating all Loopt sharing as involuntary.
Green Dot’s announcement described mobile technology, rewards, payments, and a new Silicon Valley development hub. It included costs to wind down existing services. 3
Green Dot disclosed roughly $34 million paid to Loopt stockholders and common Sequoia ownership. Sequoia partners sat on both companies’ boards. 10 Those facts warrant scrutiny, but do not establish an improper transaction. Comparing the headline price with cumulative funding also cannot calculate each investor’s return without ownership and payout terms.
The acquisition preserved value in the team and technology while ending the independent consumer product. That outcome supports a narrower conclusion than either a triumphant exit narrative or a claim that the company had no value.