Skip to content
Academy
Marketing Academy · Field Work●Human Psychology
CoreAudit· 50 minutes

Audit Your Own Funnel for Real vs. Fake Bias Cues

Warby Parker

Objective: Using your own (or a practice) product page, pricing page, and one marketing email, verify whether every anchoring, scarcity, and social-proof cue you're already running is real and defensible, then log a fix list for anything you can't verify.

You're a growth marketer at Warby Parker, the DTC eyewear company, ahead of a compliance review of the site's pricing and promotional copy.

Walk your own live pages (or a page you have edit access to) and verify three bias categories against the source data, not against how convincing the copy sounds.

Before you start

What you'll need

Free path (everything below is enough to finish)

FreeLog every cue and its verification status

Free, no account friction, works for a solo audit

The process

3 steps

Step 01 of 03

Anchoring Bias Verification

The lesson's Anchoring Bias section shows Apple projecting $999 before revealing $499 for the original iPad, then repeating the tactic with the 2024 Vision Pro at $3,499. The anchor works, but only stays inside the ethics line when the anchor number was ever a real price.

Your pricing/product page shows a 'was $X, now $Y' claim, or a highest-tier-first layout. Was $X ever the real price this exact SKU sold at, or the real un-bundled price?

Google Sheets— Pull every 'was/now' or crossed-out price on your live site into one sheet, one row per SKU.

Procedure

  1. List every product/pricing page showing a struck-through 'was' price
  2. For each, pull the actual price history from your order system or CMS changelog
  3. Mark REAL if the 'was' price was genuinely charged in the last 90 days, mark UNVERIFIED otherwise
  4. For UNVERIFIED rows, note who owns the fix (marketing copy change vs. engineering price-history feed)
Sample output
SKU        Was Price   Real Price History?   Verdict
RW-224     $310        Charged Jan-Mar 2026     REAL
RW-118     $295        No record found          UNVERIFIED
RW-330     $340        CMS default, never sold  UNVERIFIED

Healthy

Every 'was' price traces to a real price history entry within a defensible window.

Unhealthy

A 'was' price that was never actually charged, invented purely to make the current price look like a discount.

What this means

An anchor with no price history isn't persuasion, it's a fabricated number, and it's the exact 'invented inflated was price' pattern the lesson's ethics section calls out.

So what do I do about it?

SymptomActionEffort
One or more SKUs show UNVERIFIEDRoute the SKU to whoever owns pricing copy and either find the real price record or remove the strike-through30 min
YouYou can do this yourself, no engineering access required.

Step 02 of 03

Scarcity and FOMO Verification

The lesson's Scarcity section is explicit: 'if only 3 seats remain, say so; if supply is unlimited, do not pretend otherwise.' Any countdown timer or stock badge is only ethical if it reflects real, checkable data.

For every countdown timer or 'X left' badge on your funnel, does it come from a live data feed, or a static/looping asset?

Google Sheets— Open each live page in an incognito window twice, an hour apart, and log what changes.

Procedure

  1. Open each page with a countdown timer or stock badge in a fresh incognito window
  2. Note the exact time remaining or stock count shown
  3. Reload the same page after refreshing the browser (not waiting) and log whether the timer reset
  4. Cross-check any 'X left in stock' badge against the real inventory count if you have access
Sample output
Page              Cue                Reset on refresh?   Matches real inventory?
/checkout/RW-224   00:14:22 countdown   YES (reset to 15:00)  n/a
/product/RW-118    "Only 3 left"        n/a                    NO (42 units in stock)

Healthy

Timers count down in real time across refreshes; stock badges match the live inventory count.

Unhealthy

A timer that jumps back to a fixed starting value on refresh, or a stock badge that never matches inventory.

What this means

A resetting timer or a stock badge detached from real inventory is the same fake-urgency pattern the UK's 2024 DMCC Act and the FTC now penalize, not a persuasion technique worth defending.

So what do I do about it?

SymptomActionEffort
A timer resets on refresh or a stock count doesn't match inventoryEscalate to engineering to wire the cue to the real data feed, or remove it until it can bedev ticket
EitherYou or a developer can handle this, depending on your access.

Step 03 of 03

Social Proof Verification

The lesson's Social Proof section names the honest version explicitly: real review counts, 'X people bought this today' only if it's a real count, real customer logos.

Every review count, 'bought today' counter, or customer logo on your page, can you trace it to a real, named source?

Google Sheets— List every social-proof element on your top three landing pages.

Procedure

  1. List every review count, activity counter, and customer logo across your top 3 landing pages
  2. For review counts, confirm the number matches your review platform's dashboard (not a rounded-up marketing number)
  3. For customer logos, confirm each logo'd company is an active or verifiably past customer, not aspirational
  4. Flag anything you can't trace to a real source
Sample output
Element                        Source checked          Verdict
"2,340 verified reviews"        Yotpo dashboard: 2,338    REAL (rounding, acceptable)
"Trusted by 500+ brands" logo   CRM active accounts: 214  UNVERIFIED, overstated

Healthy

Every number and logo traces to a real, current data source within a reasonable rounding tolerance.

Unhealthy

A round, impressive-sounding number nobody can trace to an actual dashboard or account list.

What this means

'Trusted by 500+ brands' when the CRM shows 214 active accounts isn't optimistic rounding, it's the same overstatement the lesson warns kills trust once a customer checks.

So what do I do about it?

SymptomActionEffort
A social-proof number is more than a small rounding gap from its real sourceCorrect the copy to the real, current number5 min
YouYou can do this yourself, no engineering access required.

Final deliverable

A three-tab audit log (anchoring, scarcity, social proof) marking every live bias cue REAL or UNVERIFIED, with an owner and fix action for each UNVERIFIED row.

See a reference example
Sample output
Chewy Autoship page audit (excerpt)

ANCHORING
  "Was $54.99, Now $42.99" (Bag XL)      REAL, matched Jan invoice history

SCARCITY
  "Selling fast" badge, no countdown       UNVERIFIED, no inventory threshold defined, routed to PM

SOCIAL PROOF
  "4.7 stars, 18,900 reviews"              REAL, matches Bazaarvoice dashboard (18,873)

Success criteria

You're done when you can:

  • All three bias categories checked across at least one real or practice page
  • Every UNVERIFIED cue has a named owner and a fix action, not just a flag