Tom Tunguz argues that share of habit is a better metric for startups to focus on than engagement:
Engagement fails the majority of products as the best metric to optimize because maximizing engagement/time-on-site contradicts the product’s purpose. Google relentlessly whittles down the time it takes for users to complete a search. Sparrow reduces email client use. Online travel agents and e-commerce companies minimize conversion funnel duration. Expensify slashes time spent filing expenses.
For other products, engagement is a good metric. Facebook and Instagram and the New York Times and YouTube and other media sites do maximize time on site. The more time on site a user spends the greater the number of ad impressions and hence revenue.
But all these products seek to maximize share of habit… winning share of habit means understanding the environment of the user – what competition for time and attention does my user face – and creating a superior experience.
“Share of habit”, however, has disadvantages. You can’t measure it if you don’t have access to accurate market size data — and who has that? And it’s not a good operating metric, as it’s impacted by external factors out of your control. In that respect, it’s a vanity metric.