Performance funnel case study — Meta Ads reconciliation

01 Context

A $9 tripwire whose job was to buy a customer, not turn a profit

This was a paid freelance project for a video-maker with 20+ years behind the camera. The product was a $9 mobile-filmmaking intensive — a tripwire. A tripwire is not meant to earn: its job is to convert an anonymous click into a paying customer with a contact in Telegram, so the real money can come later from an upsell (an online course at $149, or 1-on-1 mentoring at $899).

The funnel ran end to end: Meta Ads → landing page → WayForPay checkout → Telegram bot (delivery + warm-up) → upsell.

The project ran as two launches, and my role grew between them — that growth is the spine of this case:

Act 1 — Ukraine · narrow role March 26 – April 4, 2026

The landing page and payments belonged to another contractor. My zone was tracking and running the ads: two campaigns, $253.

Ten weeks between launches — not idle: the bot was rebuilt around warm-up, the tracking was fixed.

Act 2 — Poland · full ownership delivery June 14–25, 2026 · 12 days (campaign formally stopped July 3)

Russian- and Ukrainian-speaking diaspora. I owned everything end to end — the site, the whole measurement layer, the offer, the Telegram bot, the ads. One campaign, $248.

Both launches together came to ~$501. This was a diagnostic budget with a diagnostic goal: not to scale, but to find where the funnel breaks and test the economics of the model. Every conclusion below is calibrated to that scale — 22 purchases is a signal, not a statistic, and I treat it that way throughout.

02 Hypothesis

What each launch was actually testing

Act 1 (Ukraine)

Does the tripwire mechanic work on cold traffic at all?
Which placements produce paying customers — tested as two separate campaigns (FB+IG together vs Instagram-only) rather than editing one, to keep the before/after data clean and avoid resetting Meta's learning phase.
How does the upsell sell? On Ukraine it was pitched by direct message right after purchase.

The upsell hypothesis failed: direct pitches for the big course met silence. The takeaway — a buyer has to warm up and reach out themselves, which needs a nurture sequence, not a cold pitch. That became the main change for launch two.

Act 2 (Poland)

Does the funnel carry to a new, colder, more expensive market (diaspora: higher CPM), with a setup corrected for the Act 1 lessons — Instagram-only from day one, an age focus, a clean attribution signal (strict 7-day click / 1-day view, all AI creative enhancements off, manual placements) — and a bot rebuilt around multi-step warm-up instead of direct selling?

Two further questions inside the launch, not raised on Ukraine: which placement inside Instagram (Reels / Feed / Stories) buys intent cheapest, and which of three creative angles resonates. Both were answered by campaign structure — three creatives in one ad set, letting Meta's delivery optimization decide, instead of splitting a thin budget across an A/B test.

03 Setup

A tracking stack built so campaign decisions could trust the data

Top layer: I built and owned the full stack — landing page, analytics, and conversion tracking with a CAPI safeguard (GA4, GTM, Meta Pixel + one-click Conversions API) — so decisions rested on data I could verify.

The technical layer

Static landing on Cloudflare Pages. Page weight cut from 7.4 MB to 636 KB, LCP 2.09s (green Core Web Vitals) — ready for paid traffic.
One GTM container, two mirrored receivers: GA4 for funnel analytics, Meta Pixel for the ad signal. Event funnel PageView → ViewContent → InitiateCheckout → Purchase on shared triggers. The checkout trigger fired on the payment-processor URL, not the button text, so copy edits couldn't break it.
One-click CAPI as a safeguard against lost browser signal. It paid off: on Poland, Meta's Purchase events matched the payment processor 7 = 7 — closing the measurement gap diagnosed on Ukraine.
Telegram bot (SendPulse) rebuilt: product delivery + a multi-step warm-up + engagement (chat with the author) + a buyer-source step — a third attribution layer alongside Meta and WayForPay.
UTMs via {{ad.name}} / {{adset.name}} macros — GA4 saw traffic per creative with no manual tagging. Tagging and purchase tracking were incomplete on Ukraine; fixed on Poland, which is what opened the per-creative breakdown.

GTM container: 9 tags — GA4 and Meta Pixel funnel events, both checkout tags on the shared WayForPay-URL trigger — GTM container — mirrored Meta Pixel and GA4 funnels on shared triggers. Source: GTM

Landing page hero — $9 tripwire offer 'shoot like a pro', 10 steps in 3 lessons, Telegram delivery — Landing page, above the fold. Source: slavaiordanov.com

Campaign structure

Ukraine: two Sales / ABO campaigns optimizing for InitiateCheckout — FB+IG together vs IG-only — to isolate the placement variable. Landing and payments sat with the other contractor.

Poland: Sales / ABO, optimizing for InitiateCheckout (too few purchases to optimize on Purchase directly), $25/day, broad targeting with hard limits in Controls (geo, age 22+, languages RU+UA), 3 creatives on different angles in one ad set — natural delivery optimization instead of splitting the budget on an A/B test.

04 Metrics

The numbers, split by who reported them

Spend and clicks come from Ads Manager. Payments come from WayForPay. I keep them in separate columns on purpose — the whole point of the case is what happens when you stop trusting one and start checking it against the other.

The funnel's biggest loss is not where Meta shows it: 77% of reported checkout intent never became money.

Impressions — Meta71,090

Link clicks — Meta1,264 · CTR 1.78%

InitiateCheckout — Meta96 · 7.6% of clicks

Payments — WayForPay22 · 22.9% of IC

Bar length = conversion from the previous step, on a shared 0–100% scale. The amber step is where the platform's picture and the payment processor's picture diverge.

Source: Meta Ads Manager (impressions, clicks, InitiateCheckout) vs WayForPay (settled payments) · both launches, Mar–Jul 2026

Per-campaign performance · spend & clicks: Ads Manager · payments: WayForPay
	UA · FB+IG	UA · IG-only	PL · IG-only
Role	tracking + ads	tracking + ads	full ownership
Spend	$134.96	$117.88	$248.05
Impressions	24,598	18,566	27,926
Link clicks	786	210	268
CTR (all)	5.07%	2.26%	1.82%
CPC (all)	$0.11	$0.28	$0.49
InitiateCheckout (Meta)	45	18	33
Payments (WayForPay)	9	6	7
CPA on real payments	$15.00	$19.65	$35.44
Tripwire ROAS	0.60x	0.46x	0.25x

Tripwire ROAS = revenue from the $9 product against that campaign's spend. Full-funnel ROAS (0.40x) is covered under economics below.

CTR (all) / CPC (all) count all clicks including reactions and shares; the funnel above uses link clicks only — traffic that actually reached the landing page (1.78%).

Ads Manager, three campaigns: PL $248.05, UA IG-only $117.88, UA FB+IG $134.96 — total spend $500.89, 96 InitiateCheckout; purchases recorded only on PL (7) — Ads Manager — all three campaigns, full project period. Note: purchases show only on PL (7) — the UA tracking gap is visible right in the platform. Source: Ads Manager

The headline of the whole project: Meta claimed 96 checkouts, the payment processor confirmed 22. About 77% of the platform's checkout events never became money.

Meta — checkout events reported96

WayForPay — payments confirmed22 · 23%

Bar length = absolute count of events, on a shared 0–96 scale.

Source: Meta Ads Manager (InitiateCheckout) vs WayForPay (settled payments) · both launches · UA 63→15 (24%) · PL 33→7 (21%)

Two markets, two landing pages, two stacks — the same result: ~77% of Meta's checkout events don't turn into cash. Optimizing or reporting on Meta's numbers here would have overstated the result by roughly 4×. Every decision was made on the payment processor.

Excel daily funnel — Meta InitiateCheckout beside WayForPay payments, day by day; grand total 96 IC vs 22 payments (22.9%), CPA $22.77, revenue $198 — The reconciliation itself: Meta-reported checkout intent beside confirmed WayForPay payments, day by day, per campaign. Source: Excel · META_LANDING_FUNELL_DAILY

WayForPay payments table, Polish launch — 8 successful payments and one refund; 7 of the 8 came from ads, matching Meta CAPI exactly — WayForPay payments, Poland launch, June 2026. 8 successful payments (405 UAH ≈ $9 each). Of these, 7 came from ads — Meta CAPI recorded exactly 7, so platform and processor matched (7=7). One was a direct organic purchase, outside the ad funnel. One refund (paid and reversed within minutes) is shown as a separate row. The case funnel counts only the 7 ad-attributed payments — organic sales and refunds are excluded. Source: WayForPay

Funnel economics

Break-even ROAS of the tripwire is ~1.14x at ~88% margin; actual tripwire ROAS ran 0.25–0.60x — the tripwire itself loses money, and that's by design: its job is to buy the customer.
CAC of a paying customer: $501 / 22 = $22.77 — confirmed in the payment processor.
Upsell: 0 sales from 22 buyers. On Ukraine the upsell was pitched by direct message and buyers went silent. The warm-up was rebuilt between acts, but no upsell closed before the project stopped. The upsell step is the funnel's bottleneck — and that's the main honest diagnosis of these launches.
Funnel ROAS in cash: $198 / $501 = 0.40x — below break-even. The funnel did not pay back. The diagnostic budget bought knowledge, not profit: acquisition works, monetization doesn't, and now I know exactly where it breaks.

05 Decisions

Three calls, each made on evidence I can defend

Decision 1 · Ukraine

Cut Facebook, tighten the age — on three-layer attribution, not one number

Inside the mixed campaign, Facebook looked like the star: CTR 6.72%, clicks at $0.08 (Instagram ~3%, $0.20). An age breakdown of the same campaign added a second fact: the 65+ segment produced 28 of 45 checkout events (62%) at $1.88 — 2–3.5× cheaper than any other age. Meta doesn't cross age with placement in its breakdowns, so these were two separate signals, not one story. Sales sat in WayForPay with no placement split, and Meta had already discredited its own checkouts — so the third layer settled both: I had added a source-selection step in the bot, and every buyer named Instagram. The cheap intent — by placement and by age — was hollow. A qualitative factor backed it up: the author has no Facebook presence, so FB traffic had nowhere to warm up.

Call: drop the Facebook placement, tighten age to 45, concentrate budget on Instagram. This carried into the Poland setup from day one.

Ads Manager, UA mixed campaign split by platform — Facebook: $71.50, 650 clicks, 6.72% CTR, 31 checkouts at $2.31; Instagram: $63.46, 136 clicks, 2.98% CTR, 14 checkouts at $4.53 — Meta Ads Manager, Ukraine launch — same ad set split by platform. Facebook: $71.50 spend, 650 link clicks, 6.72% CTR, $0.08 CPC, 31 checkouts at $2.31 each. Instagram: $63.46 spend, 136 link clicks, 2.98% CTR, $0.20 CPC, 14 checkouts at $4.53 each. On every platform metric Facebook looked like the better buy — cheaper clicks, cheaper checkouts, twice the intent. Every paying customer came from Instagram.

Decision 2 · Between acts

Rebuild the upsell mechanic and the tracking

From the failure of direct selling on Ukraine, the bot was rebuilt from selling to warming up — a multi-step sequence, engagement, a chat with the author, so the buyer ripens and reaches out. In parallel, the UTM tagging and purchase tracking that were incomplete on Ukraine were fixed — which is what produced the 7 = 7 match with the payment processor and a working per-creative breakdown on Poland.

Decision 3 · Poland

Placement, age and creative calls inside Instagram

Reels was the cheapest intent: 17 IC at $6.05, vs Feed 8 at $10.53 (1.74× more expensive) and Stories 8 at $7.61 → at scale, budget concentrates in Reels.
Age core 25–44 produced 26 of 33 IC (35–44 the best segment: 14 IC at $6.50). The Ukraine lesson applied pre-emptively: 55+ got under 5% of budget.
The "wow-result" creative won decisively. Inside one ad set, Meta concentrated delivery itself: the "shot-on-a-phone" video took $220 of $248 (89% of budget) and drove 32 of 33 IC at $6.88; the two other angles together got ~$28 and 1 IC. Honest caveat: the losing angles got too few impressions for a statistical verdict — the expected trade-off of running three ads in one ad set (delivery optimization, not an A/B test). One honest footnote from the ad-level data: the "losing" food close-up creative produced 1 purchase on $12.79 of spend (0.70x ROAS vs 0.25x for the winner) — a reminder that delivery optimization concentrates on checkout intent, not on purchases, and can starve a creative before it proves itself on revenue.

Inside Instagram, Reels bought checkout intent 1.74× cheaper than Feed.

Reels — 17 IC$6.05 per IC

Stories — 8 IC$7.61

Feed — 8 IC$10.53

Cost per InitiateCheckout by placement · bars scaled from zero · Source: Meta Ads Manager, PL campaign, Jun 14 – Jul 3, 2026

Ads Manager, Poland launch — ad-level breakdown: snyato-na-telefon $220.11, 32 of 33 checkouts at $6.88; the two other creatives ~$28 combined and one checkout — Meta Ads Manager, Poland launch — three creatives in one ad set, ad-level breakdown. The 'wow-result' angle (snyato-na-telefon) took $220.11 of $248.05 spend and produced 32 of 33 checkouts at $6.88 each. The two other angles got roughly $28 combined and one checkout between them. Delivery optimization concentrated the budget on its own — this was not a split test.

06 Learnings

What held up, and what I'll name as a limit

What held up

Platform metrics can't be taken on faith. Reconciling against the payment processor isn't paranoia, it's basic hygiene — the 23% pattern replicated across both launches.
Acquisition worked: 22 paying customers at $22.77 each, every one confirmed in the payment processor.
Separate campaigns beat editing a live one: clean data boundaries, no learning resets. And because the campaigns ran sequentially — the mixed one Mar 26–30, IG-only from Mar 31 — auction overlap was practically absent by construction: the second launch was the implementation of a decision, not a parallel test.
CAPI closed the tracking gap: purchases were missing in Meta on Ukraine, and matched the payment processor 7 = 7 on Poland.

What I'll call a limit, plainly

Monetization didn't close: 0 upsells from 22. The warm-up was rebuilt between acts, but 7 Poland buyers and a short window before the project stopped aren't enough to validate the new mechanic. Monetization stayed unvalidated, and I say so directly.
Too many variables changed between acts (market, landing, offer, stack, bot mechanic) — the Poland effect can't be isolated from the new-landing effect. With more budget I'd move one variable at a time.
CPA rose from Ukraine ($15–20) to Poland ($35.44): the diaspora is colder and more expensive than the home market. A fact is a fact — a Poland buyer cost roughly twice as much.
Placement-level ROAS (FB vs IG) on Ukraine can't be computed honestly: payments sat in the processor with no placement split, and there was no deduplicated CAPI. So the Facebook call rests on bot attribution, not a ROAS figure — there is no number in this case I can't defend.
Learning Limited at these budgets is an accepted state, not a mistake: exiting it would have cost more than the monthly budget.

The project was stopped by the client — workload, finances, a relocation — before the upsell step gathered enough volume to conclude.

The diagnostic goal was fully met. The monetization goal was not. Both of those are true, and the case is stronger for saying so.

07 If I'd had 10× the budget

Where the next dollar would go

Not a wishlist — a direct read of where the funnel actually broke.

The upsell step as the main object of work. The bottleneck is found; now fix it: A/B the bot warm-up sequences (length, engagement steps, chat with the author), step-by-step analytics inside the bot, and test selling the high-ticket tier via a call request instead of a payment button.
Server-side CAPI via the payment-processor callback, deduplicated by event_id (orderReference), so Meta optimizes on real payments rather than inflated checkouts.
Switch optimization to Purchase once volume reaches ~50 purchases/week; until then, InitiateCheckout as the proxy.
One variable per iteration: lock the landing and offer first, then test geo, then creative angles (developing the winning "wow-result" format in Reels) — and kill criteria for every test fixed in advance, before launch.
Systematic tripwire → upsell conversion as the core economic metric (current base 0/22 — now a baseline exists, and a rebuilt warm-up to test against it).
Standardized UTM tagging and a weekly spend → cash cohort report — the same reconcile-to-money principle, but routine instead of one-off.