One campaign, one week, one question: can this store turn a cold US visitor into a paying customer at all? Everything below is built to answer that and nothing else.
Lifetime, sales campaigns only. Benchmarks are Littledata and the 2026 Shopify cohort studies.
| Stage | NEXEDGE | Normal | Read |
|---|---|---|---|
| Cost per visitor | $3.45 | — | Too high |
| Visitor → add to cart | 4.9% | 4.6–6.0% | Normal |
| Cart → checkout | 16.7% | 30–50% | Weak |
| Visitor → sale | 0.0% | 1.4% | Zero |
| Cost per add to cart | $70.24 | target <$25 | 2.8× |
The product page already turns visitors into carts at the normal rate. That part is working. The money is leaking at the price we pay for a visitor, and again between the cart and the checkout.
| Campaign | Objective | CPM | Cost / visitor |
|---|---|---|---|
| 27072026_abo_$100/5 | Sales | $37.50 | $3.74 |
| 030826_abo_$25×4 | Sales | $149.82 | $3.55 |
| 130826_video_cbo | Sales | $189.89 | $3.43 |
| 130826_image_cbo | Sales | $142.43 | $2.92 |
Same objective on all four. One of them ran at $37.50, so the Sales objective is not what costs the money. The target for this test is a $40–$60 CPM. Not $16, and not $190.
Notice the last column. Creative has improved three times and cost per visitor has never moved: $3.74, $3.55, $3.43, $2.92. Better ads keep getting cancelled out by delivery. That is what this test changes.
Without these we are testing the same blocked page again and the $700 tells us nothing.
On a phone the popup covers 812px of an 812px screen, fires the moment the page opens, and only shows to first-time visitors. Every ad click is a first-time visitor, so all 366 paid visitors hit a full screen email wall before reading a word. It also offers 10% off on a page already showing 45% off.
It lives in Klaviyo, not in Shopify, so removing a popup in the theme does not touch it. Klaviyo account XFDdqj → Sign-up Forms.
Owner: Faiz50 × 168px, glued to the right edge at z-index 90,000, on all 15,000px of the page. On a 375px phone it eats 13% of the width and sits on top of the headlines and the comparison table.
Owner: FaizToday it is free over $99 and $5.95 below, which the buyer first meets at checkout. On a $49 order that is a 12% surprise, and cart to checkout is already 16.7%. The 60-day bundle at $95 also misses the free shipping line by $4.
Owner: Faiz · decision PatrickCreator seeding orders raised as Shopify draft orders fire real Purchase events into the pixel. Those are the only Purchase events the pixel has ever seen, which is part of why Purchase optimisation has nothing true to learn from.
Owner: FaizOne campaign. One ad set. Four ads. Nothing else in the account running.
Keep the campaign objective on Sales. Change the conversion event inside the ad set from Purchase to Add to cart. Two different settings at two different levels.
Why: Meta has to find people likely to do the exact event we name. The pixel holds zero real buyers, so on Purchase it has nothing to copy and it guesses, which is wide and expensive. It holds 18 real carts and earns more at a normal 4.9% rate, so on Add to cart it has something to learn from.
Add to cart is only offered when the objective is Sales. If the campaign is switched to Traffic or Engagement the option disappears. Checked on 10 Aug: this pixel is not caught by Meta’s health and wellness restriction, so the event is selectable. Many supplement advertisers are blocked from it. We are not.
| Setting | Value |
|---|---|
| Objective | Sales · conversion location Website |
| Performance goal | Maximise number of conversions |
| Conversion event | Add to cart on pixel 3204156606640319 |
| Structure | 1 campaign · 1 ad set · 4 ads |
| Budget | $100/day × 7 days = $700, or $70/day × 7 = $490 if capped at $500 |
| Placements | Advantage+ minus Audience Network |
| Audience | Broad US · men · 30–55 |
| Attribution | 7-day click, 1-day view |
| Destination | nexedge.store/products/nexedge-natural-testosterone-support |
These four made 13 of the 18 lifetime carts and all 3 checkouts. Two video, two image.
| Ad | Spend | Carts | Checkouts | Cost / visitor |
|---|---|---|---|---|
| NOT READY FOR TRT | $135.45 | 4 | 2 | $2.51 |
| 3.1 — Didn’t Go Back To Nothing | $113.20 | 5 | 0 | $2.46 |
| I DIDN’T TELL HER I STARTED TAKING ANYTHING | $120.89 | 2 | 1 | $2.37 |
| WIFE NOTICED AD 2 | $174.20 | 2 | 0 | $2.72 |
Four is the cap, not a starting point. The 13 Aug launch put 11 ads on $113, which is about $10 each, and no ad-level number in it could be read. At $100/day, four ads is already thin.
The hard part is leaving it alone.
$700 at a $50 CPM should buy roughly 350 to 450 visitors.
Judge this on cost per purchase. Never on cost per cart. Cost per cart will look excellent and mean nothing, because we asked Meta for carts and Meta is very good at finding people who add to cart and leave.
| What we see | What it means | Next |
|---|---|---|
| 3+ sales | The store converts. We have a business to tune. | Go Switch to Purchase goal, raise AOV, then build a pre-sell page |
| 1–2 sales | Alive but thin. | Hold One more week at the same settings |
| 0 sales · carts <$25 · cart→checkout near 40% | Traffic and page are fine. The break is in the checkout or the payment gateway. | Stop ads Test a real purchase end to end on a phone |
| 0 sales · cart→checkout still <20% | Cart friction, or the traffic really is cart-adders only. | Stop ads Open a cart on a phone and watch what happens |
| Carts still over $50 | Delivery never recovered from the restarts. | Stop The account may need a fresh ad account, not a fresh campaign |
It is not a profit attempt. At $49 with roughly $31 contribution, nothing at this price is profitable even if it works perfectly. If anyone scores this on return on ad spend, a successful test will look like a failure and get switched off.
It is not the growth plan either. The two levers that make this business work are a normal cost per visitor and a higher order value: the 90-day bundle at $135 keeps about $97 against $31 on a single bottle. Both come after we have a yes. There is also a strong case for sending ads to a pre-sell page instead of the product page, which is how Mars Men and Menerals both run. That is the next build, once we know the store can close.
Right now we are buying one thing: a yes or a no.