Discount Strategy & Unit Economics: How to Run Promotions Without Destroying Your Margins

Here's a number that should make every DTC founder uncomfortable: the average ecommerce brand discounts 30-40% of total revenue. That's not a promotion strategy — that's a margin hemorrhage disguised as growth.

We've managed over $500K/month in ad spend across 100+ brands at ATTN Agency, and the pattern is always the same. A brand launches, grows on full-price sales, then hits a plateau. Someone suggests a 20% off sitewide sale. Revenue spikes. Everyone celebrates. Then the next month is flat, so they run another sale. Within a year, 40% of customers have never paid full price, and the brand is wondering why their unit economics fell apart.

Discounting isn't inherently bad. But undisciplined discounting is a death spiral. Let's break down exactly how promotions impact your unit economics — and how to run them without torching your margins.

The True Cost of a Discount (It's Not What You Think)

Most operators think a 20% discount means a 20% hit to revenue. That's wrong. It's a much bigger hit to profit.

Here's a simple example:

• Product price: $100
• COGS: $35
• Contribution margin before discount: $65 (65%)
• Price after 20% discount: $80
• COGS (unchanged): $35
• Contribution margin after discount: $45 (56%)

Your revenue dropped 20%, but your contribution margin dropped 30.8%. That's because discounts come entirely out of margin — COGS doesn't move. The higher your COGS as a percentage of price, the more devastating discounts become.

At 50% COGS (common in food/beverage), a 20% discount wipes out 40% of your contribution margin. At 70% COGS (commodity products), a 20% discount eliminates 66.7% of margin. At that point, you're paying to acquire customers who aren't profitable on first order and may never be.

The Volume Multiplier Problem

The math gets worse. To maintain the same gross profit dollars after a discount, you need significantly more volume:

| Discount | Margin Before (65%) | Volume Increase Needed | |----------|---------------------|----------------------| | 10% | → 55% | +18.2% | | 15% | → 50% | +30.0% | | 20% | → 45% | +44.4% | | 25% | → 40% | +62.5% | | 30% | → 35% | +85.7% |

A 25% off sale requires you to sell 62.5% more units just to break even on gross profit. And that's before factoring in the incremental fulfillment, shipping, and customer service costs that come with higher volume.

Ask yourself: does your supply chain, warehouse, and support team scale that cleanly? For most DTC brands under $20M, the answer is no.

How Discounts Warp Customer Lifetime Value

The margin hit on individual orders is bad enough. But the real damage is what discounts do to customer behavior over time.

Discount Conditioning

Data from brands we work with shows a consistent pattern: customers acquired during a promotion expect promotions. Their repeat purchase behavior becomes discount-dependent.

Here's what we typically see:

• Full-price acquired customers: 35-45% repeat purchase rate, average 2.8 orders in Year 1
• Discount-acquired customers: 20-28% repeat purchase rate, average 1.9 orders in Year 1
• Deep discount acquired (30%+): 12-18% repeat purchase rate, average 1.4 orders in Year 1

The LTV gap is enormous. A full-price customer with a $100 AOV, 40% repeat rate, and 2.8 orders generates roughly $280 in Year 1 revenue. A discount-acquired customer with an $80 AOV (they bought on sale), 24% repeat rate, and 1.9 orders generates $152.

That's a 46% LTV reduction from a single promotional acquisition strategy. And it compounds — those discount-trained customers are also more likely to churn after their first purchase if they don't see another offer.

The Unsubscribe-to-Discount Pipeline

We've audited email flows for dozens of DTC brands, and one pattern is universal: brands that lead with discounts in email see 2-3x higher unsubscribe rates on non-promotional emails.

You're literally training your list to ignore you unless you're offering money off. That's not a customer relationship — it's a Pavlovian pricing experiment that you're losing.

The Five Discount Archetypes (And Their Unit Economics Impact)

Not all discounts are created equal. Here's how the major discount strategies stack up from a unit economics perspective.

1. Sitewide Percentage Off (The Blunt Instrument)

What it is: 20% off everything, Black Friday-style.

Unit economics impact: Maximum margin destruction. Every order — including ones that would have converted at full price — takes the hit. Based on our data, 40-60% of revenue during a sitewide sale comes from customers who would have purchased anyway.

The cannibalization math: If 50% of sale revenue would have happened at full price, and you discount 20%, your true incremental margin on the sale is deeply negative. You gave away 20% on half your revenue to earn the other half at reduced margins.

When it works: Almost never, unless you're liquidating inventory that's going to zero value anyway.

2. Tiered Spend Thresholds (The AOV Play)

What it is: Spend $100, get 10% off. Spend $150, get 15% off. Spend $200, get 20% off.

Unit economics impact: Significantly better than flat discounts because you're anchoring the discount to higher cart values. If your baseline AOV is $85 and customers stretch to $150 for the 15% tier, you're collecting $127.50 versus $85 — a 50% AOV increase for a 15% discount.

Key metric: Track the "stretch rate" — what percentage of customers move to a higher tier versus buying at their normal level. Good execution sees 30-40% of customers stretching at least one tier.

When it works: Brands with multiple SKUs at different price points, where adding a second or third item is natural.

3. Gift With Purchase (The Margin Protector)

What it is: Spend $X, get a free product/sample/gift.

Unit economics impact: This is the most margin-friendly promotion type. If your gift costs $5 in COGS and your customer perceives it as $25+ in value, you've created a 5:1 perceived discount ratio. Compare that to a percentage off where the ratio is always 1:1.

Real example: A skincare brand we work with tested 20% off ($18 discount on $90 AOV) against a free deluxe sample ($4.50 COGS, $22 retail value). The GWP offer converted 12% better and preserved $13.50 more margin per order. Over a 10,000-order campaign, that's $135,000 in saved margin.

When it works: Brands with high perceived-value products that have low marginal COGS (beauty, supplements, food).

4. New Customer Only Discounts (The Acquisition Tax)

What it is: First order 15% off, welcome offer, etc.

Unit economics impact: Better than sitewide because existing customers don't get the discount. But you're still starting every new customer relationship in the red. The critical question is payback period — how many orders until you've recovered the acquisition discount plus CPA?

Framework: Calculate your discounted first-order contribution margin, subtract CPA, and determine how many repeat orders at full margin you need to break even.

Example:

• First order: $100 AOV × 15% discount = $85 revenue, $50 contribution margin (vs. $65 full price)
• CPA: $40
• First order profit: $10 (vs. $25 at full price)
• Lost margin from discount: $15 per customer
• At scale (10,000 new customers/month): $150,000/month in margin given away

Is that $150K/month buying you proportionally more volume? For most brands, a 10-15% welcome offer increases conversion rate by 15-25%. Do the math for your specific numbers.

5. Loyalty/VIP Discounts (The Retention Lever)

What it is: Exclusive offers for repeat customers, loyalty program perks, early access pricing.

Unit economics impact: The best ROI of any discount type, because you're rewarding already-proven customers. A customer with 3+ orders has an 80%+ chance of ordering again — they don't need a discount to convert, but a small one can increase frequency.

Smart execution: Don't discount the products they already buy. Discount adjacent products to increase category penetration. A customer buying your hero SKU doesn't need 10% off that SKU — they need a reason to try your new product line.

Building a Discount Strategy That Doesn't Kill Your Margins

Here's the framework we use with DTC brands to structure promotions around unit economics, not revenue vanity metrics.

Step 1: Set Your Margin Floor

Before any promotion, define the minimum acceptable contribution margin per order. For most DTC brands, this should be:

• Healthy: 55%+ contribution margin
• Acceptable during promotions: 40-50%
• Red line: Below 35%

If a promotion structure pushes you below your red line, it doesn't run. Period. Revenue is not profit.

Step 2: Segment Before You Discount

Different customers deserve different offers. Stop treating your entire list like one audience.

| Segment | Discount Strategy | Max Discount | |---------|-------------------|-------------| | Never purchased | Welcome offer (one-time) | 10-15% | | Purchased once, 30-60 days ago | GWP or free shipping | $0-5 COGS | | Purchased 2+, active | VIP early access, new products | 10% on new items only | | Lapsed (90+ days, 1 order) | Win-back with threshold | 15% on $X+ spend | | Lapsed (90+ days, 2+ orders) | Aggressive win-back | 20% (they've proven value) | | High-LTV VIP | Exclusive bundles, experiences | Don't discount — add value |

Step 3: Track Promotional vs. Full-Price Revenue Mix

This is the single most important metric most brands aren't tracking. Your promotional revenue ratio should stay below 25% of total revenue. Above 30%, you're becoming a discount brand whether you want to be or not.

Calculate monthly:

• Total revenue from orders with any discount applied
• Divided by total revenue
• = Promotional revenue ratio

If this number is climbing quarter over quarter, you have a structural discounting problem.

Step 4: Measure Discount ROI Like You Measure Ad Spend

Every promotion should have a target ROAS — return on discount spend. Calculate it:

Discount ROAS = Incremental profit from promotion ÷ Total discount dollars given

"Incremental" is the key word. You need to estimate what revenue would have been without the promotion (use prior non-promotional periods as a baseline) and only count the net-new revenue.

Most brands find their actual discount ROAS is 0.5-1.5x. Meaning they're spending $1 in discounts to generate $0.50-$1.50 in incremental profit. Compare that to paid media ROAS of 3-5x and suddenly the "free" promotion looks expensive.

Step 5: Time-Box Everything

Open-ended discounts are margin killers. Every promotion needs:

• Start date and hard end date (48-72 hours max for urgency)
• Inventory cap (first 500 orders, limited quantity)
• Customer cap (one use per customer, no stacking)

Scarcity drives conversion without extending the margin damage window. A 48-hour flash sale at 20% off does far less long-term damage than a week-long promotion at 15% off, even though the per-order discount is higher.

The Discount Calendar: Frequency Matters

How often you promote matters as much as how deep you discount. Here's what we recommend:

• Maximum 4-6 promotional events per year for a premium DTC brand
• 8-10 max for a mid-market brand
• Monthly promotions = you're a discount brand, restructure your pricing instead

Map your promotional calendar to natural purchase cycles and cultural moments. If your replenishment cycle is 60 days, a promotion at day 45 accelerates repurchase without training discount dependency. A promotion at day 20 just cannibalized a full-price order.

The Promotional Dead Zone

Never run promotions in the 2 weeks following a major sale. Customers who just bought don't need a reason to feel buyer's remorse, and customers who missed the sale need to learn that full price is normal.

Real Numbers: Discount Strategy Overhaul Case Study

A supplement brand we work with was running 20% off sitewide promotions monthly. Here's what happened when we restructured their approach:

Before (monthly 20% off sitewide):

• Average monthly revenue: $420K
• Promotional revenue ratio: 52%
• Blended contribution margin: 41%
• Monthly contribution profit: $172K

After (segmented, 4x/year promotions):

• Average monthly revenue: $385K (yes, it dipped)
• Promotional revenue ratio: 18%
• Blended contribution margin: 58%
• Monthly contribution profit: $223K

Revenue dropped 8%. Contribution profit increased 30%. The brand was more profitable on less revenue because they stopped giving margin away to customers who would have bought anyway.

Month 6 post-change, revenue recovered to $410K as full-price conversion rates improved. Customers stopped waiting for sales.

Common Discount Mistakes (And What to Do Instead)

Mistake: Discounting to hit revenue targets. Do instead: If you need to hit a revenue number, increase ad spend on proven campaigns. At least that investment compounds into customer acquisition.

Mistake: Matching competitor discounts. Do instead: Compete on value, not price. If your competitor is at 30% off, they're likely bleeding margin. Let them.

Mistake: Using discounts to fix retention. Do instead: If customers aren't coming back, the product or experience has a problem. A discount on a mediocre experience just creates a cheaper mediocre experience.

Mistake: Stacking discounts (loyalty + sitewide + welcome). Do instead: One offer per customer per transaction. Stacking is how $100 orders become $55 orders with $12 in margin.

Mistake: Not tracking discount-acquired cohort LTV separately. Do instead: Split your LTV analysis by acquisition channel AND offer type. You'll likely find that full-price acquired customers are worth 2-3x discount-acquired ones.

The Bottom Line

Discounts are a tool, not a strategy. Used surgically, they accelerate inventory movement, reward loyal customers, and create urgency. Used carelessly, they erode margins, condition customers to wait for sales, and turn premium brands into commodity players.

The brands winning in DTC right now aren't the ones with the biggest sales — they're the ones with the discipline to protect their margins while their competitors race to the bottom.

Run the numbers. Set your margin floor. Segment your customers. And stop giving away profit to people who were going to buy anyway.

Your CFO will thank you.