Twine: the anti-retirement calculator

The category convention — fifteen form fields, a giant scary total, and a polite suggestion to "consult an advisor" — fails on its own terms. It demands work the user isn’t equipped to do, then rewards them with a number they don’t know how to use.

We started from the opposite premise. Users want two things: can I retire at all, and what do I do about it. The interface should be designed for those two answers, not for a goal-setting calculation.

The team had two weeks to prove or kill the thesis. I led divergent exploration in the first days, narrowed on a quiz-first format, and built a high-fidelity Figma prototype ready for testing in the second week.

1. 01
   Divergent exploration

   Wide scan of category conventions and adjacent visualizers for paradigms worth borrowing.

2. 02
   Format convergence

   Settled on a quiz-first input model — short, conversational, opt-in to depth — rather than the standard up-front form wall.

3. 03
   Prototype

   High-fidelity Figma prototype with the three outcome variants, ready for moderated head-to-head testing.

We ran the prototype against NerdWallet’s retirement calculator with target users in the second week. Users consistently responded better to “you can retire at 67” plus actionable advice than to a six-figure goal they had no plan to reach.

The advice itself was segmented into three buckets, each with appropriate language and a different call to action.

Twine the company shut down before the visualizer reached production. What didn’t shut down was the approach. The same principle — verdict over total, action over arithmetic, empathy at the bottom of the funnel — has held up across every consumer financial product I’ve touched since.

The app itself was an Apple App of the Day three times, held a 4.6 App Store rating, and managed $20M in AUM at its peak — credentials carried by the team and the craft, even when this particular feature didn’t ship.