April closed at $428.6K revenue, up 12% month over month. Good number. Wrong conclusion, if you stop there.

The tempting story is that acquisition got stronger because sessions were up 18% and paid search looked healthy. The numbers don't support that. Returning customers contributed $175.7K — up 38% from March — and bigger baskets added $23.6K. Conversion barely moved (+0.1pts). New-customer revenue was almost flat at $252.9K (+3% MoM). One discount-led collection (Weekend Sets) made the topline look cleaner than the economics underneath it.

The real read: the store monetised existing demand better, but did not materially improve first-purchase efficiency. Returning customers and AOV carried the month. Paid search harvested intent that brand and email had already created. The acquisition engine did not get stronger — it got busier.

That distinction matters for May. If returning-customer demand softens, the topline has no fallback. The store needs to fix cold-start conversion before it can count on growth that doesn't depend on re-engagement.

Two things improved at once, but only one deserves the headline.

Returning-customer revenue grew 38% MoM ($175.7K vs $127K). New-customer revenue grew 3% ($252.9K vs $255K). People already in the system came back, spent more, and lifted the month without the store solving cold-start conversion.

The returning lift came from three places:

• Email-driven repeat purchases — the April 8 restock email generated $28.4K from 1,840 clicks. 15.4% click-to-purchase rate on a segment that had already bought once. The subject line ('Back in stock: the pieces you browsed') referenced specific product interest, not a generic sale.

• Loyalty-program redemptions — 340 members redeemed points in April, up from 210 in March. Average redemption order: $112 vs $86.40 sitewide. Loyalty customers self-select into bigger baskets.

• Paid search brand campaigns — Google Brand Exact returned 7.2x ROAS on $12,400 spend. Efficient, but not acquisition. Re-engagement with a media budget.

Paid search deserves credit for harvesting intent. It does not deserve credit for transforming acquisition economics. Brand and high-intent search worked. Prospecting traffic (Google Shopping broad, Meta Prospecting LAL 2%) mostly made the dashboard busier — $31.2K spend, 1.6x ROAS combined.

One sentence: the store made more money from people it had already convinced, and the acquisition channels that look healthy are mostly catching demand that email and loyalty already created.

The surprising bit: the healthiest-looking growth lever is not the one I'd scale first.

Weekend Sets posted $77.9K revenue — second-highest collection. On the surface, a win. The economics say otherwise:

• 22% average discount rate — deepest in the store, driven by a 'Buy 2, Save 25%' promo that ran 18 of 30 days

• 42% gross margin — 16pts below Core Layers (58%), 13pts below Everyday Essentials (55%)

• 29% returning-customer share — lowest in the top cohort. The collection depends on attracting new buyers who need a discount to convert

• $34.20 first-order CPA via paid channels — 40% above the $24.40 store average

The collection that looks like a winner on revenue is one of the least attractive pieces of growth once you care about economics. It bought topline with margin and paid acquisition subsidy.

Meanwhile, the boring collections are the real engine:

• Core Layers: $92.4K revenue, 58% margin, 8% discount, 46% returning share

• Everyday Essentials: $81.7K revenue, 55% margin, 12% discount, 44% returning share

• Denim Staples: $63.3K revenue, 57% margin, 10% discount, 48% returning share — highest repeat rate in the store

Not 'what sold.' What sold in a way you'd want to repeat. Weekend Sets sold. You would not want to repeat the economics that made it possible.

The leak is specific. That's good news — it means the fix is specific too.

Weekend Sets put up $77.9K revenue, but needed a 22% average discount rate to do it and finished with the weakest gross margin in the top cohort at 42%. That is not an evergreen growth engine.

What the discounting actually cost:

• $10.1K in forgone margin — at the collection's natural margin (estimated 55% without the promo), Weekend Sets would have contributed $42.8K in gross profit. At actual 42% on discounted sales, it contributed $32.7K. The 'Buy 2, Save 25%' promo cost $10.1K in gross profit to generate $77.9K in topline.

• First-order CPA inflation — paid channels sending traffic to Weekend Sets averaged $34.20 CPA, 40% above the $24.40 blended average. The discount attracted lower-intent buyers who needed both a deal and a paid ad to convert.

• Cannibalisation signal — Denim Staples and Core Layers both saw returning-customer share dip in weeks 4-6, overlapping with the Weekend Sets promo window. Repeat buyers shifted purchases to the discounted collection instead of buying at full margin.

The sturdier pattern came from Core Layers and Everyday Essentials: lower discounting, better repeat mix, stronger margin retention. Less exciting in a Slack screenshot. More valuable in an operating review.

Revenue flatters both stories. Margin tells you which one deserves more capital. Right now, the collection getting the most promotional support returns the least per dollar of cost.

The April read sets up a specific tension for May.

If returning-customer demand holds, the store clears $440K in May — modest growth driven by the same engine (email, loyalty, brand search). Two risks:

1. Seasonal dip in repeat demand. April benefited from a spring restock cycle that drove email clicks. May has no equivalent product event unless you create one. Without a restock or new-drop trigger, returning-customer revenue softens 10-15%.

2. Weekend Sets promo hangover. The 'Buy 2, Save 25%' promo ran 18 days in April. Customers who bought at 25% off resist paying full price in May. If the collection reverts to standard pricing, Weekend Sets conversion drops 20-30%. If the promo stays, margin stays compromised.

If the store does nothing differently, May comes in at $410-420K — a step back from April — with margin improving slightly (less promo dilution) but topline looking weaker. The team reads that as 'May was soft.' It was not soft. It was April's discounting pulling demand forward.

The better path: accept a smaller topline in May, protect margin on Weekend Sets (cap discounts at 12%, run a targeted email-only offer instead of sitewide), and invest the recovered margin into testing new-customer acquisition creative for Core Layers. That collection converts repeat demand well — the question is whether it converts cold traffic at acceptable CPA. If it can, the store has a second growth engine. If it cannot, at least you know the ceiling before spending against it.

Three moves, in priority order.

1. Split every weekly review into new vs returning revenue. The business could tell itself an acquisition success story that the numbers don't support. Add a row: new-customer revenue, returning-customer revenue, blended. If returning share drops below 35%, that is a warning signal — not a dashboard decoration.

2. Scale the boring winners first. Core Layers ($92.4K, 58% margin, 46% returning share) and Everyday Essentials ($81.7K, 55% margin, 44% returning share) are the healthiest units in the month. They deserve more merchandising attention, more email featuring, and a $5-8K paid-search test to see if they convert cold traffic. If they acquire at or below $24.40 blended CPA, you have a scalable, high-margin growth channel. That is worth knowing.

3. Put a hard economic guardrail on discount-led growth. Weekend Sets stays in the mix, but only with a margin floor (no collection below 48% gross margin without CFO sign-off) and a clear reason for the subsidy. 'It drove revenue' is not a reason. 'We are clearing end-of-season inventory to free warehouse space for the summer drop' is a reason. One is a strategy. The other is a habit.

The point is not to kill the revenue winner. The point is to stop confusing loud growth with good growth. Loud growth shows up in the topline. Good growth shows up in the margin, the repeat rate, and the acquisition efficiency. This month, those three metrics tell different stories. Trust the margin.

Three questions I'd ask Anna next. Each one follows from something this analysis found.

1. Decompose my revenue change this month into traffic volume, conversion rate, and average order value. Which component contributed the most?

If AOV drove 52% of the gain (as it did here), the operating question becomes 'what raised basket size?' — not 'what drove traffic?' Different question, different levers, different budget decisions.

2. Which collections are driving revenue but dragging margin because of discounting? Show me the discount rate vs gross margin scatter for all collections.

The scatter makes it visible: the collection in the top-right (high discount, low margin) looks like a winner on revenue but borrows from next month's economics. Once you see the scatter, you cannot unsee it.

3. Compare new vs returning customer revenue over the last 90 days. Who is carrying growth, and where is acquisition spend underperforming?

If returning customers carry 41% of revenue and grow faster than new, the acquisition budget needs a different conversation — not 'spend more' but 'spend on what converts first-purchase buyers at or below blended CPA.'

Paste any of these into Anna after connecting Shopify. The answer will not be a chart. It will be a direction.