50Queen · Launch Ad Set · Funnel Readout

$319 bought 228 leads and nothing else

Everything below is the launch ad set alone — US · Women 45–65 · Broad · Lead, live from 21 August, paused 24 August. The advertising works and the quiz works. Payment is where it stops: three women got as far as a payment form and none of them finished.

Ad set 120252198044700131 21 Aug 17:42 → 25 Aug 01:13 UTC Paused $80/day Validation campaign excluded

Where the money went

Three and a half days of delivery at $80/day. Everything above the paywall is healthy or better. Everything below it is zero.

Spend

$318.64

3.5 days · $80/day

Link clicks

1,006

$0.32 each · 7.70% CTR

Leads

228

$1.40 each

Paywall views

237

24% of quiz entries

Checkout attempts

$106.21 each

Customers

$0.00 revenue

The one sentence version

You are paying market-beating rates for attention — a 7.70% link click-through rate at $0.32 a click, and a $1.40 lead — and then losing 98% of it at the moment of payment. No amount of ad optimisation fixes a funnel that converts nothing at the end.

Ad performance

The ad set holds two creatives. One of them received essentially all the budget.

Delivery and cost, launch ad set — note CTR/CPC here are all-click metrics
Ad	Spend	Impr.	Reach	Freq.	Clicks	CTR	CPC	CPM
A56 · Says Who · 15s	$311.67	12,854	10,933	1.18	1,380	10.74%	$0.23	$24.25
A40 · Says Who 40s · 15s (starved)	$3.05	42	39	1.08	3	7.14%	$1.02	$72.62
Ad set total	$318.64	13,072	11,050	1.18	1,402	10.73%	$0.23	$24.38

What the 1,402 “clicks” actually were

Meta's headline clicks metric counts every click on the ad unit, not just the ones that open your site. Only 72% of them were link clicks; the rest were reactions, saves, comments, shares, caption expands and taps on the page name. Read the link-click line, not the click line.

Click composition
Metric	Count	Share of clicks	Meaning
Clicks (all)	1,402	100%	Every click on the ad unit
Link clicks / outbound	1,006	71.8%	Actually opened the site
Unique link clicks	940	67.0%	Distinct people who opened it
On-platform clicks	396	28.2%	Reactions, saves, shares, page taps
— of which post reactions	48	—	Likes and other reactions
— of which saves	10	—	Saved the post
— of which comments / shares	3 / 3	—	Public engagement

Conversion down the ad funnel

Link clicks to landing to lead to checkout
Stage	Count	Rate from previous	Cost each
Impressions	13,072	—	$0.024
Link clicks	1,006	7.70%	$0.32
Unique link clicks (people)	940	93.4%	$0.34
Landing page views	904	96.2%	$0.35
Leads (email captured)	228	25.2%	$1.40
Checkout attempts (with a client-side event)	3	1.3%	$106.21
Purchases	0	0.0%	—

Video engagement

How much of each creative actually gets watched
Ad	Impressions	50% watched	95% watched	Reached 95%	ThruPlays
A56 · Says Who · 15s	12,854	5,090	2,866	22.3%	2,798
A40 · Says Who 40s · 15s	42	11	4	9.5%	3

Two things in this data

The creative is working. A 7.70% link click-through rate at a $24 CPM, with 22% of impressions watching a 15-second video to 95%, is a strong ad — and the extra 396 on-platform clicks are a second, free signal that the creative resonates. Frequency is 1.18, so the audience is nowhere near fatigued; you could spend considerably more before delivery degrades.

The in-app browser is not costing you much. 940 distinct people clicked the link and 904 landing views fired — 96.2%. Against raw link clicks (1,006, including the same person tapping twice) it is 89.9%. Either way the handoff into the Facebook browser is working normally; the loss is roughly 36 people, about $12, and is not worth engineering time.

Your creative test never ran

A40 took $3.05 of $316 — 42 impressions, 3 clicks. Meta concentrated delivery on A56 within hours and you have no read at all on the 40s-targeted cut. If you want that answer it needs its own ad set with its own budget; inside a shared one it will never get delivery.

Quiz drop-off, every screen

Distinct funnel ids reaching each screen since the ad set went live, in the order she meets them. Bar length is volume relative to the first screen. Two screens lose almost everything; the other twenty-seven lose almost nothing.

Question Interstitial — no answer required Conditional branch — only some see it

s1-1Fitness that works with your body, not against it971—

s1-2You're not unfit419−56.9%

s1-3How old are you?393−6.2%

s1-4A year from now, what would you most want to be true?382−2.8%

s1-5What's changed in the last few years?369−3.4%

s1-6It's physiology. Not willpower.360−2.4%

s2-1In a typical week, how much do you move?351−2.5%

s2-2Can you get up from the floor without using your hands?344−2.0%

s2-3Can you carry shopping up a flight of stairs without stopping?337−2.0%

s2-4Can you stand on one leg for 10 seconds?335−0.6%

s2-5By mid-afternoon, how do you usually feel?334−0.3%

s2-6Where do you feel stiff or achy in the mornings?332−0.6%

s2-7These aren't separate problems293branch

s2-8Do you ever leak when you cough, sneeze, laugh or jump?331−0.3%

s2-9Then we won't ask you to jump256branch

s2-10Anything we should work around?330−0.3%

s2-11Mobility first, then strength326−1.2%

s3-1How are you sleeping?325−0.3%

s3-2Are hot flushes or night sweats part of your day?3250.0%

s3-3Who else are you looking after at the moment?3250.0%

s3-5What have you got at home?324−0.3%

s3-6Three days. Fifteen minutes. A chair.320−1.2%

s4-1When you've tried to get moving before, what got in the way?3200.0%

s4-2Fifteen minutes is not selfish253branch

s4-3What matters most about staying strong?308−3.8%

s4-4What should we call you?301−2.3%

s4-5Where should we send your plan?285−5.3%

s5-1Putting your plan together235−17.5%

s5-5Paywall236—

Read this shape, not the numbers

The bar collapses once, at the top, and then stays almost perfectly flat for twenty-five screens. 971 women see question one and 419 answer it. Of those 419, 56% go all the way to the paywall. Nobody is bored by your quiz — it holds attention better than most onboarding flows do, losing under 2% per question through the middle.

The second real leak is the email gate: 285 reach "Where should we send your plan?" and 235 come out the other side, so 50 women complete a 25-question quiz and then refuse to give an address. That is the most expensive 18% in the funnel, because every one of them was already qualified.

Answer rate on the questions that record one

Screens that fire a submit event convert at 95–100% — she answers essentially every question she is shown. The blanks are interstitials with only a Continue button, or the free-text name and email screens, which do not emit the event.

Viewed versus answered
Screen	Viewed	Answered	Rate
s1-1 · First question	971	419	43.2%
s1-3 · How old are you?	393	382	97.2%
s1-4 · A year from now…	382	369	96.6%
s1-5 · What's changed…	369	353	95.7%
s2-4 · Stand on one leg	335	333	99.4%
s2-8 · Do you ever leak…	331	330	99.7%
s3-2 · Hot flushes…	325	325	100%

Note what this rules out. The intimate questions — incontinence, hot flushes, who you care for — convert at 99–100%. Sensitivity is costing you nothing. The only screen with an answer-rate problem is the first one.

Paywall event analytics

All paywall events in the ad set's window, split by pricing arm. The experiment went live mid-afternoon on 23 August, so the 114 pre-experiment visitors are this ad set's earlier traffic, who saw the old $99 / $45 / $16 ladder before assignment existed.

Paywall funnel by pricing arm, distinct people
Arm	Offer	Viewed	Scrolled	Scroll rate	Abandoned	Checkouts	View → checkout
founding49	$99 → $49 once	67	33	49.3%	34	3	4.5%
weekly8	$99 anchor, $16 → $8/wk	56	33	58.9%	24	0	0.0%
pre-experiment	$99 / $45 / $16	114	56	49.1%	46	0	0.0%
All	—	237	122	51.5%	104	3	1.3%

How far down the page she gets

People reaching each scroll depth, by arm. Every checkout came from the shortest-read arm.

founding49

weekly8

pre-exp.

25%50%75%100%

The counter-intuitive bit

weekly8 is read more thoroughly and bought less. 45% of its scrollers reach the bottom of the page against 27% for founding49, yet founding49 produced all three checkouts and weekly8 produced none. Reading further is not interest — it is looking for the catch. An $8-a-week price anchored against $99 invites her to work out what the total is; $49 once needs no arithmetic.

At 67 versus 56 people this is nowhere near significant and you should not call the experiment on it. But it is the only directional signal you have, and it points the same way as every checkout that has ever opened.

Half of them never scroll at all

115 of 237 paywall viewers produce no scroll event whatsoever. They land on the price and leave without moving the page. Combined with 104 explicit abandon events, the paywall is not losing an argument — it is not getting one started.

Every checkout this ad set produced

Stripe livemode, filtered to real people. Four sessions across three consecutive days — but only three have a matching client-side event, so only three are evidence that a woman actually tried to pay. All on the $49 one-time plan, all founding49, all arriving through the Facebook in-app browser. None of them paid.

Real user checkout sessions
Opened (UTC)	Email	Amount	Location	Device	Checkout type	Outcome
23 Aug 15:45	howellshari3@…	$49	Toronto	Android · FB app	Hosted	Expired unpaid
24 Aug 03:05	sunseeker67@…	$49	Winnipeg	iPhone · FB app	Hosted	Open, unpaid
25 Aug 01:13	haldanetara09@…	$49	Edmonton	iPhone · FB app	—	No checkout event
25 Aug 03:19	nadinedesroches08@…	$49	Toronto	Android · FB app	Embedded	Started, closed in 0.93 s

Everything else in the Stripe account within this window is noise and is not counted: four sessions from your own Hong Kong IPs, and one from a curl/8.7.1 scraper.

A Stripe session is not proof she tried to pay

haldanetara09 has a live $49 Stripe session with her email on it — and no checkout event of any kind. Her last recorded action is landing on the paywall; the session appears four seconds later with no click, no scroll, and no checkout_screen_viewed in between.

haldanetara09 — last recorded seconds
Time (UTC)	Event
01:13:18.320	`paywall_viewed`
01:13:19.165	`paywall_section_viewed` — last event, ever
01:13:23	Stripe session created — no click recorded

There is no checkout_started for her either, client-side or server-side. So either her analytics died while the page kept working, or something creates sessions without a deliberate click. Until that is resolved, Stripe session count overstates real intent.

The new checkout fires its event server-side, losing the join

Seven checkout_started events exist in total. The two most recent — the only ones from the embedded flow — carry no $session_id and no quiz_public_id, and their device_type is lowercase mobile/desktop with no browser or geo, where every client-side one has Mobile, Chrome, Barrie.

Every checkout_started ever fired
When (UTC)	Session id	Funnel id	Who
21 Aug 09:56	✓	tLlQc0Jq…	Founder e2e
23 Aug 15:45	✓	KeVZY0lW…	howellshari3
24 Aug 03:05	✓	dWtAT7t_…	sunseeker67
24 Aug 10:36	✓	oOZTjH6O…	Founder test
25 Aug 00:54	✓	oOZTjH6O…	Founder test
25 Aug 03:19	null	none	nadinedesroches08
25 Aug 03:26	null	none	Founder e2e

Consequences worth fixing before the next spend: a checkout_started from the embedded flow cannot be opened as a session replay, cannot be attributed to a pricing arm through the funnel id, and cannot be tied back to its Stripe session. Four of the seven are your own tests, so filtering this event without excluding them roughly doubles the apparent demand.

What the sheet-close data does and doesn't say

Every embedded checkout that has opened has also closed — but the dwell times say these are not all the same event:

Embedded checkout sheet, all opens to date
Who	Opened	Open for
Your own test (HK)	00:54:50	11.6 s
nadinedesroches08	03:19:27	0.93 s
Your e2e test	03:26:37	13.5 s

Your two tests held the sheet open for 11–13 seconds, which reads as looking at it and dismissing it deliberately. So checkout_sheet_closed is an ordinary dismissal event, not evidence of a crash. Only nadinedesroches08's 0.93 s is anomalous — an order of magnitude faster — and that is a single observation which could equally be a mis-tap outside the sheet. Worth reproducing; not yet a finding.

Separately, and still true: Apple Pay is not a Stripe payment_method_type — it is a wallet carried on card, so ["card"] does not by itself disable it. What the embedded sessions did drop is link and amazon_pay, and payment_method_configuration_details is null where the hosted sessions carried pmc_1RebPq….

The read

The ad is not the problem and neither is the quiz. A 7.70% link click-through rate, a $0.32 link click, a $1.40 lead and a frequency of 1.18 are numbers most advertisers would take happily — and the audience is not close to saturated. A quiz that loses under 2% per question across twenty-five questions, including questions about incontinence, is a well-built quiz. Both are assets.

The economics do not work yet and are not close. Three checkout attempts on $319 is $106 each. Even if every one had paid — and none did — that is $106 spent to earn $49. The gap is not a tuning problem; it is roughly 2x, and that is the optimistic reading where checkout converts at 100%. At a realistic 30% it is 7x.

What I would do, in this order

Watch these four replays first. These are the four women with a live Stripe session — but only three fired a checkout event. Two reached the hosted page, one started the embedded flow and closed it in under a second, and haldanetara09 fired nothing at all. Watching them is the fastest way to learn which of those is a bug.

Session replays — the four who attempted payment
Who	When (UTC)	Checkout	Replay
howellshari3@	23 Aug 15:42	Hosted	01a02f49…
sunseeker67@	24 Aug 02:57	Hosted	01a031b4…
haldanetara09@	25 Aug 01:09	No event	01a03676…
nadinedesroches08@	25 Aug 03:10	Embedded	01a036e5…

Reproduce on a real phone, and fix the instrumentation. Walk the funnel in the Facebook in-app browser and tap “Start my plan”. Two things to settle: whether the sheet can dismiss itself, and how a Stripe session gets created with no client-side event. Until the second is answered, session count overstates intent.
Make the first screen respond instantly. 552 women lost on one tap is the biggest single number in this document. The page is one 487 KB document — 263 KB over the wire even compressed, carrying 110 KB of JavaScript, 73 KB of CSS and all 29 screens inline. On a mid-range Android in the Facebook browser that parses for a noticeable beat after first paint, and taps landing in that window get swallowed. This is a hypothesis, not a finding — but it is testable in an hour and it is the one plausible mechanical explanation for a 57% drop on a screen whose successors convert at 97%.

What would change the verdict

One completed purchase. Everything here is a funnel with a zero at the end, and zero divided by anything is still zero. The moment a single stranger pays $49, every ratio above becomes a real conversion rate you can optimise instead of a hypothesis.