A product can change course; its users may not follow
A pivot either keeps the people and changes the product for them, or keeps the product idea and starts over with new people. Trying both usually fails.
You can pivot a product, but not a user base.
In August 2010 Digg released a rebuilt version of its site that changed how stories reached the front page and removed features its heaviest users relied on. Those users protested by flooding the front page with links to Reddit, a rival. Traffic fell in the months that followed, and many of the site’s most active members moved elsewhere for good. Digg had changed its product. Its audience declined to change with it.
Gary Chou, who runs the network of people across Union Square Ventures’ portfolio companies in New York, puts the lesson in ten words. The word pivot has spread through startup circles over the past two years, largely through Eric Ries’s writing, and it usually describes the product: a new market or a new model. The line points at what that framing leaves out. Code can be rewritten in a month. The people using it came for something specific and bring expectations that do not update when the company does.
The pivots that worked mostly accepted this. Odeo was a podcasting company; when its founders turned to Twitter in 2006, they did not try to convert podcast listeners into tweeters. They started with a new audience. Flickr began as a feature inside an online game and grew into a photo site with a different crowd. Burbn, a check-in app, became Instagram last October by stripping itself down to photos and launching nearly from scratch. In each case the team, the know-how and some of the code carried over. The users mostly did not.
Where the line bends is when the users lead. Twitter’s own history has cases of this: people began marking topics with the hash sign and addressing each other with the at sign before the company built either into the product. A change that follows what users already do is not really a pivot away from them. It is a pivot toward them, and users will follow a product that is catching up with their habits.
It also bends when the change serves the same need better. Customers who used a product to solve a particular problem will usually accept a new approach to the same problem, as long as it works. What they resist is being asked to want something else.
The practical reading is to decide, before the change, what is being kept. If the users are the asset, every change has to make their lives better, and the company should ask them before it acts. If the team or the technology is the asset, it is often cleaner to treat the pivot as a new launch, tell the existing users plainly, give them a way to take their data with them, and accept that most will leave. Trying to keep both by moving an audience somewhere it did not choose to go is how Digg lost its own.