Watch what users actually do rather than what you built for, then cut everything except the single behavior they love and rebuild the entire product around it.
Instagram is the photo-sharing app that started life as a cluttered location check-in product called Burbn and became a cultural giant within two years. The through-line across its stories: Kevin Systrom and Mike Krieger watched what users actually loved, killed everything else, and rebuilt around the single behavior that worked, then let a following-based social graph do the rest.
Pivots: unbundling the one feature people actually used
The problem. Systrom raised a $500,000 seed round in early 2010 and built Burbn, an HTML5 location app where users could check in, make plans, earn points, and post photos. It was cluttered and undifferentiated, competing directly with Foursquare and Gowalla in a crowded check-in field, with no single thing it did better than anyone else.
The approach. After Krieger joined in March 2010, the two studied Burbn's usage data and found people were really using it for one thing: sharing photos. They stripped the product down, cut the check-ins and points, and rebuilt around three things only (photos, filters, and likes/comments), renaming it Instagram, a portmanteau of "instant camera" and "telegram."
How it solved it. The focused native iOS app took roughly eight weeks to build and shipped on October 6, 2010. Killing the me-too check-in product and keeping only the loved behavior turned a struggling Foursquare clone into the top free photo app on launch day, a pivot so clean it is now taught as the textbook case of unbundling a beloved feature.
PMF: instant, undeniable fit before the product was even finished
The problem. Consumer apps usually have to fight for early traction, and Instagram launched into a photo category already crowded with camera and filter apps like Hipstamatic. There was no guarantee that a two-person team's stripped-down app would find demand fast enough to matter.
The approach. Instagram married Hipstamatic-style filters to a social feed: filters made mediocre phone photos look good, and the feed made sharing them feel rewarding. Systrom seeded the app pre-launch to a small group of designers and creatives with large Twitter followings, whose first posts set a high aesthetic bar.
How it solved it. Fit was immediate. Instagram registered 25,000 users on launch day (October 6, 2010), roughly 100,000 in the first week, and 1 million users by December 2010, about ten weeks in. It reached around 27 million registered users by early 2012, all before an Android app existed, a demand curve that leaves little doubt about product-market fit.
Virality: every photo became an outbound ad
The problem. A two-person startup with $500,000 in seed money had no budget for paid user acquisition, and a photo app is only worth opening if there is fresh content and people to see it. Instagram needed growth that compounded on its own.
The approach. Instagram built cross-posting directly into the sharing flow, letting users push every photo to Twitter and Facebook with one tap. Each shared image carried the app's distinctive filtered look and a link back, turning users into a distribution channel across networks that already had hundreds of millions of people.
How it solved it. Growth was overwhelmingly organic at near-zero customer acquisition cost. The seeded creators' filtered posts acted as free, high-taste advertising, and cross-posted photos on Twitter and Facebook functioned as outbound ads that pulled in new signups, which is how the app climbed to 1 million users in ten weeks without a marketing spend.
Network Effect: the following graph that Facebook chose to buy
The problem. A photo feed is only interesting if the people you care about are posting, so Instagram's value depended entirely on accumulating the right users and the connections between them. That same dynamic, if it took hold, would make the product very hard for a competitor to dislodge.
The approach. Instagram used an asymmetric following/follower model rather than Facebook's mutual friending, so a person could amass a huge one-to-many audience without reciprocation. As users, content history, and creator audiences accumulated, the social graph itself became the durable asset, the engine that later powered influencers, Stories, and Reels.
How it solved it. The graph became a moat too expensive to attack head-on. In April 2012 Facebook acquired Instagram for roughly $1 billion (about $300 million cash and $700 million in stock) when the company had 13 employees, around 30 million users, and zero revenue, moving over a single weekend after Twitter had reportedly offered around $500 million. Zuckerberg was buying the habit and the community, not the technology.
Founder-Market Fit: a founder wired into the exact taste-making network the product needed
The problem. An aesthetic-driven photo app lives or dies on who posts first, because the earliest content sets the cultural tone and the quality bar that attracts everyone after. A generic launch to a generic audience would have produced a generic feed.
The approach. Systrom, a Stanford graduate who had worked at Google and at Odeo (the startup that became Twitter), used those connections to hand-seed the app to designers and creatives who already had large Twitter followings. He was targeting exactly the high-taste creators his product needed, a network he happened to be embedded in.
How it solved it. Those seeded creators' first posts became free, high-taste advertising that defined Instagram's look and drew in followers, feeding the launch-day surge to 25,000 users and 1 million within ten weeks. The founder's specific network was the launch strategy: a different founder without that access could not have manufactured the same instant credibility.