6-IG

Before Instagram there was Burbn, and Burbn was doing badly. Kevin Systrom built it in 2010 as an HTML5 web app for people who liked whiskey and going out. You could check in to bars and restaurants, post plans for later, earn points for hanging out with friends, and — among a long list of other things — upload photos. It was Foursquare-adjacent, feature-rich, and utterly undifferentiated. Systrom raised half a million dollars on it and brought on Mike Krieger. Then the two of them did the thing almost nobody does. They looked hard at how the small user base actually behaved, and found that people were essentially ignoring the check-ins, the plans, and the points. The only feature getting real use was photos. So they took Burbn apart. They cut everything except photo sharing, comments, and likes. Systrom has described looking at competitors in the photo space and concluding that the two things standing between phone photography and mass adoption were that photos took too long to upload and that they looked bad. So they solved exactly those two problems: filters that made a mediocre phone photo look intentional, and an upload flow that started transferring the image while the user was still typing the caption, so posting felt instantaneous. Everything else was thrown away. Eight months of work, deleted. Instagram launched on October 6, 2010, and got roughly 25,000 users on the first day. A million within two and a half months. In April 2012 — with thirteen employees and no revenue — Facebook bought it for approximately $1 billion. It is now one of the largest media platforms in the world and effectively invented the profession of the visual influencer. Three things about this case are worth separating from the mythology. The pivot was subtraction, not addition. Guy calls it the ERASER. The instinct under pressure is to add features until something works. Systrom and Krieger did the opposite: they removed until only the used part remained. The product got dramatically smaller and dramatically more valuable at the same moment. The usage data existed because they shipped. They did not deduce that photo sharing was the opportunity. They observed it, in the behavior of a modest number of real users on a mediocre product. That data has no substitute — no amount of planning would have generated it. Burbn’s entire value was as an instrument for measuring what people wanted. Speed was a feature, not an implementation detail. The decision to start uploading before the user finished the caption is a small engineering choice that changed the emotional experience of posting from “submit and wait” to “done.” Instagram won partly because it felt fast, and that came from taking a boring problem seriously. Burbn was a bad product that succeeded at one job: it told its founders which of its eight features mattered. That is what shipping is for. The lesson: launch the cluttered version, then watch. Your users will tell you which one feature is the company