Headspace

A meditation and mental-wellness app, founded 2010 by Andy Puddicombe and Richard Pierson. Headspace is the most-cited single product in this wiki’s onboarding/SaaS-conversion coverage — it shows up as a subject of three distinct A/B-tested wins across two sources, which is enough to promote it from a case study to a recurring reference point.

The three A/B tests

SourceChangeReported lift
I Studied 1,460 Onboarding Flows (video)Personalization quiz allows multi-intent goals (pick more than one reason for meditating) instead of one+10% free-trial conversion
Copy These SaaS Growth Tricks (video)Renamed “free trial” to “30-day guest pass”+7% signups
Copy These SaaS Growth Tricks (video)14-day trial on annual plan vs 7-day trial on monthly”Double-digit” conversion lift

What’s notable is that all three are copy and offer-shape changes, not product changes. Headspace doesn’t ship new mechanics for these wins — they reshape the framing layer around an existing offer. The pattern is a clinic in how much conversion sits on top of a stable core product.

The biases behind the lifts

LiftUnderlying bias
Multi-intent goalsPersonalization that “earns its keep” — users come with more than one pain point (see Onboarding flow)
“Guest pass” renameFraming effect — connotation > denotation
Longer trial on annualAnchoring bias — trial length as the comparison anchor

Two of the three trace cleanly to Daniel Kahneman’s prospect-theory lineage. The first is an HCI/onboarding pattern that doesn’t fit the prospect-theory frame but composes well with it.

Why Headspace surfaces so often

Tim Gabe’s commentary in Copy These SaaS Growth Tricks (video) is roughly: Headspace is unusually willing to A/B test the subscription frame itself rather than the product. Many B2C subscription apps treat the offer as fixed and iterate the marketing page; Headspace iterates the offer’s naming, duration, and intent capture, which is where most of its compounding wins come from.

The aggregate moral, from across both sources: subscription conversion is mostly an exercise in framing. The trial mechanics haven’t moved; the names and shapes have.

Sources