You can increase margin without touching a single price tag, because for most stores the margin killers aren’t product costs — they’re marketing costs per order and discount depth. Three levers move margin while prices stay fixed: lower the acquisition cost baked into each order by getting more orders from owned channels, cut the average discount by reserving codes for moments that need them, and raise average order value with relevant cross-sells. All three run on the same machinery: behavioral email and SMS automation. This article walks through each lever with numbers, then the exact flows that pull them.

The margin squeeze nobody itemizes

Your P&L probably looks something like this: healthy gross margin on paper, thin profit at the bottom. Between the two sits a growing wedge of ad spend, discount codes, and shipping promos — costs that attach themselves to every order but never show up on a product page. Revenue grows 20% year over year, profit doesn’t. If that pattern sounds familiar, why more revenue can still mean less ecommerce profit unpacks it in detail.

Raising prices feels like the obvious fix, and sometimes it’s right. But in a niche where customers compare tabs before buying, a 10% price increase can cost you the conversion rate that makes your ads work at all. So most owners quietly rule it out and go back to squeezing the ad account. There’s a third path between “raise prices” and “spend more on ads”, and it’s the one this article is about.

Why the usual moves don’t restore margin

More ad spend buys revenue at declining efficiency — the extra orders come in at a higher cost per order than the ones you already have, which drags margin the wrong way. More promotions buy revenue by giving margin away directly; a store running a permanent 15% welcome code plus monthly sales has effectively cut prices without the honesty of saying so. And cost-cutting on the product side (cheaper packaging, slower shipping) saves single-digit euros while risking the reviews that keep conversion up. None of these touch the actual leak: what it costs you, in ad spend and discount, to generate each order.

The margin math per order

Illustrative example — swap in your own numbers. A store sells at an average order value of EUR 70 with 60% gross margin, so EUR 42 gross profit per order before marketing. Now subtract the order’s share of costs:

• Blended ad cost per order: EUR 16
• Average discount given (blended across all orders): EUR 6
• Contribution left: EUR 20 per order

Move two dials without changing the price. First, shift enough orders to owned channels that blended ad cost per order drops from EUR 16 to EUR 13. Second, tighten discounting so the blended average falls from EUR 6 to EUR 4. Contribution rises from EUR 20 to EUR 25 — a 25% profit-per-order improvement, invisible to the customer. Add a modest AOV lift from cross-sells and it compounds. These figures are invented to show the mechanics, but the structure of the calculation is exactly what you should run on last month’s data.

Three levers, in working order

1. First: cut the acquisition cost inside each order. Build abandoned cart and post-purchase flows so repeat and recovered orders stop consuming ad budget. This lever is the biggest and the fastest, and it’s the core of growing ecommerce profit without constantly increasing ad spend.
2. Second: discount with intent instead of by default. Remove blanket codes. Keep incentives for two moments only — a first-purchase nudge for subscribers who haven’t bought within 14 days, and a win-back offer for customers inactive 90+ days. Everyone else gets relevance, not discounts.
3. Third: raise AOV with automated cross-sells. Recommend the accessory, refill, or companion product at the moment it makes sense — after purchase, not as a popup mid-checkout. Don’t bother yet with loyalty programs, bundles engineering, or subscription models; those are quarter-two projects once the flows run.

The automations, specified

Post-purchase cross-sell (the AOV and repeat lever)

• Trigger: order delivered.
• Segment: all customers, branched by product category bought; exclude anyone with an open unshipped order.
• Timing and channel: email at day 5 with a usage tip (builds the open habit, sells nothing); email at day 14 recommending one or two products bought together with theirs, full price, no code.
• Content: category-specific. Bought a coffee grinder? Show filters and beans, with one line on why they pair. Not a generic bestseller grid.
• Goal: second orders at zero ad cost and zero discount.

Win-back with a controlled incentive (the discount-discipline lever)

• Trigger: no purchase for 90 days.
• Segment: past customers only; exclude anyone who received a discount in the previous 30 days.
• Timing and channel: email at day 90 with new arrivals since their last order — no discount; email at day 105 with a single, expiring incentive if margin allows, sized to your numbers (free shipping often costs less than 10% off); SMS on the final day of the offer for contacts with consent.
• Content: reference what they bought, show what’s changed. The scarcity is real because the code actually expires.
• Goal: reactivate lapsed customers while confining discounts to the one segment where they’re genuinely needed. This discipline matters — selling more without training customers to wait for discounts covers the behavioral side.

Numbers to track monthly

• Contribution per order: AOV minus product cost, minus blended ad cost per order, minus blended discount per order. The headline metric for this whole effort.
• Blended discount rate: total discounts given divided by gross revenue. Watch it fall as blanket codes disappear.
• Repeat purchase rate: customers with a second order within 90 days.
• Average order value, split by first orders versus repeat orders — cross-sells should lift the repeat side first.

Running this in Omnisend

The flows above are platform-agnostic; I happen to run them in Omnisend. Before settling on it I tested Klaviyo and Omnisend side by side on my own stores and chose Omnisend — for me it was easier to self-manage without an agency, and the pricing left more of the margin this article is trying to protect. For these specific levers, the useful parts are the product recommendation blocks in post-purchase emails (they pull “bought together” data from your store), discount codes generated per-flow with real expiry dates, and segment exclusions so a customer never stacks the win-back code on top of another offer. Fair warning: the cross-sell logic depends on clean product and order data, and if your catalog has three near-duplicate variants of everything, tidy that before trusting the recommendations. And none of this repairs margin if the underlying unit economics are broken — automation multiplies a working business, it doesn’t fix a losing one.

Your next step

Run one calculation this week: last month’s total discounts given, divided by gross revenue. If your blended discount rate is above 5% and you can’t say precisely which segments earned those discounts, start there — kill one blanket code and move the incentive into a win-back flow with an expiry date.