Every action a user takes during a free trial increases their likelihood of converting to paid. Not because the product gets better — because the user gets more invested.
This is the sunk cost effect applied to trial design. And the research behind it is remarkably consistent.
The Psychology of Sunk Cost
Arkes and Blumer (1985) established the foundational finding: people who invest money, effort, or time into something become psychologically committed to continuing — even when the rational calculation says stop.
In their theater ticket study, customers who paid full price for season tickets attended significantly more plays than those who received a discount. Same plays, same seats. The only difference was how much they'd invested.
The core mechanism: "the desire not to appear wasteful" — a deeply ingrained social norm. Brain imaging studies show the dorsolateral prefrontal cortex (involved in implementing social rules) activates during sunk cost decisions.
The IKEA Effect: When Labor Leads to Love
Norton, Mochon, and Ariely (2012) at Harvard Business School ran four studies showing that personal labor dramatically inflates perceived value.
In the IKEA box experiment, people who assembled a storage box themselves were willing to pay 63% more for it than people given an identical pre-assembled box.
In the origami experiment, creators valued their amateurish origami as highly as expert-made origami. Labor inflated perceived value to the point of delusion about quality.
But here's the critical boundary condition: the IKEA effect requires successful completion. When participants built something and then destroyed it, the valuation effect collapsed. Incomplete or destroyed work doesn't generate the effect.
The implication for trials: users must FINISH something. A half-built FAQ system doesn't create investment. A completed 3-topic FAQ sprint does.
The Endowed Progress Effect: Give People a Head Start
Nunes and Dreze (2006) ran one of behavioral economics' most elegant experiments. Customers at a car wash received loyalty cards:
- Group A: 8-stamp card, starting from zero
- Group B: 10-stamp card, with 2 stamps already filled in
Both groups needed exactly 8 purchases to earn a free wash. The only difference was perceived progress.
Results: Group B had a 34% redemption rate vs. 19% for Group A. That's 79% higher completion from artificial progress alone. Group B also completed faster — the time between purchases decreased as they approached the goal.
One critical caveat: the effect disappeared when no reason was given for the head start. The reason could be completely arbitrary — "special promotion" — but it had to exist.
Effort Investment Beyond Money
Cunha and Caldieraro (2009) showed that sunk cost effects extend to purely behavioral investments — not just money. Through an effort-justification mechanism, people account for the behavioral resources they've invested. If asked to spend time selecting a product, they value the chosen product MORE to justify the effort.
This means even small effort investments trigger the effect:
- Filling out a profile
- Customizing settings
- Recording one FAQ answer
- Uploading a logo
Each small action adds to the psychological investment stack.
Navarro and Fantino (2005) even demonstrated sunk cost effects in pigeons — this behavior is deeply rooted in behavioral patterns, not just sophisticated cognitive reasoning.
How Investment Accumulates During a Trial
The investment builds in layers:
Layer 1 — Setup Investment (Day 1): Account creation, profile customization, tool configuration. Endowed progress effect kicks in — each completed step moves the progress bar forward.
Layer 2 — Content Investment (Days 2-7): Creating first FAQ topics, recording answers, building a knowledge base. The IKEA effect activates — the content they create becomes THEIRS. They value it disproportionately.
Layer 3 — System Investment (Days 8-14): Workflow integration, team sharing, link embedding. Full sunk cost escalation — too much invested to walk away without loss.
Layer 4 — Identity Investment (Ongoing): "I'm someone who builds systems." "We're the kind of team that documents things." The investment becomes self-identity — and people don't abandon their identity easily.
The Conversion Benchmarks
Industry data from First Page Sage (86 SaaS companies, Q1 2022-Q3 2025) shows the sunk cost principle in action:
- Opt-in trials (no credit card): 18.2% average conversion
- Opt-out trials (credit card required): 48.8% average conversion
The mere act of entering a credit card creates a sunk cost. That 2-3x difference comes largely from commitment investment, not from filtering out tire-kickers.
Onboarding checklist data from Userpilot (188 SaaS companies, 2024) adds more context:
- Average checklist completion: 19.2% (median just 10.1%)
- Checklists with gamification increase completion 20-30%
- A 25% increase in activation leads to a 34% revenue boost
Most trials fail at the investment-building stage. The onboarding doesn't create enough sunk cost to sustain engagement past the initial enthusiasm.
Action vs. Inaction Framing
A 2018 study (four experiments, mini meta-analysis d = 0.37) found that framing escalation as ACTION and de-escalation as INACTION increases the sunk cost effect.
This means:
- "Continue building your system" (action frame) beats "Keep your subscription" (passive frame)
- "Your 3 FAQ topics will be archived" (loss of action) beats "Trial expired" (neutral)
- Every day of the trial should feel like an active choice to invest more
The Emotional Driver
A PLOS ONE study across three experiments found that the sunk cost effect is mediated by emotional response — not just cold calculation.
People FEEL the pull of their investment. The negative emotion of potentially wasting what they've built drives continued engagement more than any logical argument.
But there's a counterintuitive finding: when people explicitly JUSTIFY their decisions, they become MORE resistant to the sunk cost effect. Let users feel their investment through progress indicators and celebration. Don't make them rationalize it.
The Ethics of Investment Design
Sunk cost effects are technically a cognitive bias — they lead to "irrational" decisions by definition. There's a genuine ethical tension in deliberately designing for them.
The ethical framework: the investment users make must create GENUINE value. The FAQ content they build should actually serve them. The system should work for them beyond the trial period. The sunk cost should align with real utility — they'd lose something genuinely valuable, not manufactured switching costs.
The goal isn't to trap people into paying for something they don't use. It's to help them build something valuable enough that paying to keep it is the obvious choice.
The Nervous System Connection
Investment behavior maps onto autonomic states:
- Effort investment activates dopamine reward circuits — each completed step creates positive anticipation
- Visible progress signals safety — "I'm building something real" engages the ventral vagal system
- Loss framing triggers threat response — protecting what you've built activates sympathetic fight-or-flight
- Identity investment supports vagal regulation — "I'm someone who builds systems" creates regulatory capacity
Effective trial design balances dopamine-driven investment building with ventral vagal safety — creating momentum without triggering defensive responses that lead to cancellation.
Key Statistics
- 63% more willingness to pay for self-assembled products (Norton et al. 2012)
- 34% vs 19% completion from artificial progress alone (Nunes & Dreze 2006)
- 48.8% vs 18.2% conversion — opt-out vs opt-in trials (First Page Sage)
- 19.2% average onboarding checklist completion (Userpilot 2024)
- 20-30% boost from gamified checklists (Userpilot 2024)
- 25% activation increase → 34% revenue boost (Fairmarkit)
- d = 0.37 effect of action framing on escalation (2018 meta-analysis)
Post #164 in the Fleshtimer research series. Research document: sunk-cost-escalation-trial-investment.md