Nudge or Sludge? Auditing Warby Parker's Real Checkout Flow
Objective: Browse Warby Parker's real checkout flow live, identify which of the four choice-architecture mechanisms (defaults, loss framing, anchoring, social proof) it uses, and flag anything that crosses from a legitimate nudge into sludge.
You're a CRO analyst asked to benchmark a well-regarded DTC checkout before your own team redesigns theirs. Rather than trust a summary, you're auditing the real, live flow yourself.
Add a pair of glasses to cart, walk the checkout to the final screen without paying, and log every choice-architecture technique you can find using the lesson's own four-mechanism list as your checklist.
Before you start
What you'll need
Free path (everything below is enough to finish)
Free, no account needed beyond an existing Google account
The process
2 steps
Step 01 of 02
The lesson lists four classic nudge mechanisms: defaults (pre-ticked boxes), loss framing ('don't lose' beats 'save'), anchoring (the first number distorts later judgment), and social proof (stated behavior of others).
Walking Warby Parker's real site from product page to final checkout screen, which of these four mechanisms actually appear, and where exactly?
Procedure
- Add a frame to cart and start checkout in a private/incognito window
- Log every screen in a sheet with columns: Screen, Mechanism, Exact Copy/Element
- Note any pre-checked boxes (insurance add-on, email opt-in, extras)
- Note any price anchor (a 'compare at' price, a bundle savings callout)
- Note any loss-framed or social-proof copy on the way to payment
Screen | Mechanism | Exact Element Cart | Anchor | 'Bundle & save $50' next to full-price total Cart | Default | Blue-light lens upgrade pre-selected Shipping | Social proof | none found Checkout | Loss framing | none found
Healthy
At least 2 of the 4 mechanisms are present, and each one maps to a real, visible element you can screenshot.
Unhealthy
You log a mechanism that isn't actually there because you expected to find it, this is the confirmation-bias trap the audit is meant to catch.
What this means
A real audit finds gaps as often as hits. Warby Parker not using every lever everywhere is itself a data point about what they consider worth the friction.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| A default pre-selects a paid add-on with no visible way to decline in the same screen | Flag it for the ethics step next, a hidden decline path is a sludge risk | 5 min |
Step 02 of 02
The lesson defines sludge, Thaler's term, as choice architecture that benefits the architect at the chooser's expense: fake countdowns, hidden opt-outs, confirmshaming, roach-motel cancellation flows.
For every mechanism you logged, is the underlying claim true and the opt-out visible in the same screen? Which ones pass, and which ones would you flag to legal?
Procedure
- For each logged mechanism, write whether the underlying number/claim is true (a real price, a real stated fact)
- Check whether declining any pre-selected default takes one click or requires hunting
- Mark each row Nudge (passes both checks) or Sludge (fails either)
- Write one sentence recommending a fix for any row marked Sludge
Mechanism | Claim True? | Opt-out visible? | Verdict Bundle & save $50 anchor | Yes, compare price matches site-wide list price | N/A | Nudge Blue-light lens pre-selected | N/A | Yes, one-tap remove in same screen | Nudge
Healthy
Every mechanism you found passes both checks, or you have a specific, one-sentence fix for the one that doesn't.
Unhealthy
You mark something Sludge without checking whether the claim is actually true first, guessing instead of verifying is the same shortcut that creates sludge.
What this means
The nudge/sludge line is not about the technique, defaults and anchors are neutral tools. It's about whether the claim is true and the exit is easy.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| An add-on default has no visible one-click removal | Move the decline control into the same screen as the default, not a settings page two clicks away | dev ticket |
Final deliverable
A logged audit of Warby Parker's checkout mechanisms with a Nudge/Sludge verdict and one fix recommendation.
See a reference example
Nykaa checkout audit (excerpt) Screen | Mechanism | Element | Claim True? | Opt-out visible? | Verdict Cart | Anchor | 'MRP 1,999, you pay 1,299' | Yes, matches product page MRP | N/A | Nudge Cart | Default | 'Nykaa Prime trial' pre-checked | N/A | No, removal buried in order summary footer | Sludge Fix: move the Prime trial checkbox decline control next to the checkbox itself, not the footer.
Success criteria
You're done when you can:
- Logs at least 3 real, verifiable mechanisms from the live site
- Correctly verifies claim-truth and opt-out visibility before assigning a verdict
- Provides one concrete fix for any row marked Sludge