Ad Spend vs Discounting: Which Grows Profit Faster

Two different growth levers, one budget

Ad spend and discounting both aim to grow revenue, but they work through different mechanisms — ad spend buys additional traffic that wouldn't have found the store otherwise, while a discount converts more of the traffic you already have (and, for existing customers, can pull forward purchases that would have happened anyway at full price). Comparing them fairly means asking the same question of both: for a given dollar of incremental investment, how much incremental profit does it produce, after accounting for who would have bought anyway.

The discount side has a hidden cost that's easy to underweight: every customer who would have paid full price and bought anyway now pays less, with zero incremental volume to show for that portion. This is sometimes called discount leakage or cannibalization — the discount applied to already-committed buyers, as opposed to the discount that actually persuaded a new or marginal buyer to convert. Any fair comparison to ad spend needs to separate the leakage portion from the incremental portion, because ad spend (when properly measured) is close to 100% incremental by construction, buying traffic that wasn't already coming.

Worked example: a storewide 15% discount

Assume a Shopify store normally does $20,000/week in revenue at 55% gross margin ($11,000 gross profit), and runs a storewide 15% discount for a week. Assume the discount lifts unit volume by 25% (a reasonably strong response) and that 60% of buyers during the discount week would have bought anyway at full price (leakage), while 40% are incremental buyers persuaded by the discount. New revenue: original 20,000 units-equivalent x 1.25 volume x 0.85 price factor — expressed in dollars, $20,000 x 1.25 x 0.85 = $21,250. New COGS: at 45% of original full-price revenue-equivalent, cost per unit is unchanged even though price dropped, so COGS scales with volume, not price: original COGS was $9,000 (45% of $20,000); at 1.25x volume, COGS = $11,250. Gross profit = $21,250 - $11,250 = $10,000, actually $1,000 lower than the non-discounted week's $11,000, despite higher revenue.

Worked example: the same dollar spent on ads instead

Assume instead the store keeps full price and spends $1,500 (roughly the revenue given up in the discount scenario, for a comparable resource commitment) on incremental Google Shopping and Meta traffic at a 3.0x ROAS. Incremental revenue = $1,500 x 3.0 = $4,500, at the normal 55% gross margin = $2,475 gross profit, minus the $1,500 ad spend = $975 net incremental profit for the week, on top of the store's normal $11,000 base gross profit — total $11,975, versus $10,000 in the discount scenario. In this worked comparison, the ad spend path outperforms the discount path by $1,975 for the week, driven almost entirely by the discount's leakage cost on already-committed buyers.

When discounting can outperform ad spend

This doesn't mean discounting is always worse — it depends heavily on the leakage assumption. If leakage were much lower, say 20% of discount-week buyers would have bought anyway rather than 60% (realistic for a discount targeted only at cart-abandoners or a specific lapsed-customer segment rather than storewide), the math flips: less of the discount is wasted on committed buyers, and more of the volume lift is truly incremental. Segmented, targeted discounts (a win-back email discount to lapsed customers, a first-purchase discount only shown to new visitors) structurally avoid most of the leakage problem that storewide discounts carry, because the audience receiving the discount is, by construction, less likely to have bought at full price.

Worked example: a targeted discount with low leakage

Assume the same 15% discount but offered only via a win-back email to customers who haven't purchased in 6+ months, a segment worth $2,000/week in potential full-price revenue if fully reactivated, with an assumed 15% leakage (most of this segment genuinely wasn't going to buy without a nudge) and a 30% volume lift among responders. Revenue from this segment: $2,000 x 1.30 x 0.85 = $2,210. COGS at 45%: original $900 at base volume, at 1.30x volume = $1,170. Gross profit = $2,210 - $1,170 = $1,040, compared to a non-discounted baseline for that segment of $2,000 x 0.55 = $1,100 if they'd bought at full price — but since most of this segment wasn't buying at all without the nudge, the realistic comparison is against $0 baseline profit from a segment that would otherwise be inactive, making the $1,040 close to fully incremental.

Blending both levers instead of choosing one

In practice, the more useful question usually isn't 'ads or discounts' but 'which discount structure and which ad spend allocation, together, produce the most total incremental profit.' A workable split: use storewide discounts sparingly and only around genuine demand events (where the ad-spend comparison above suggests they underperform on pure profit but may serve other goals like inventory clearance or press coverage), and reserve targeted, segmented discounts for low-leakage use cases like win-back, first-purchase incentives, or bundling — while treating incremental ad spend as the primary lever for pure profit growth when ROAS is comfortably above break-even.

The break-even ROAS a discount needs to clear

A useful sanity check for any discount decision: what ROAS-equivalent would the same dollar amount need to hit if spent on ads instead, to match the discount's actual incremental profit? In the storewide 15% discount example, spending the same $1,500-equivalent revenue-given-up on ads at breakeven (1.0x, meaning $1,500 ad spend returns exactly $1,500 revenue, or roughly $825 gross profit at 55% margin) would already outperform the -$1,000 discount result. That means the storewide discount, in this worked scenario, underperformed even a mediocre ad campaign — a useful reality check before assuming a sale is automatically good for the business.

Inventory clearance changes the calculation entirely

The comparison above assumes normal, sellable inventory with a stable cost basis. If the actual goal is clearing aging or excess inventory that's tying up cash or approaching obsolescence, the profit-per-dollar comparison to ad spend isn't the right frame — the relevant comparison becomes discounted-sale-now versus continuing carrying cost, markdown risk, or eventual write-off. In that specific case, a discount that looks unprofitable against an ad-spend comparison can still be the right call, because the alternative isn't 'sell at full price via ads' but 'don't sell this inventory at all in a reasonable timeframe.'

Measuring your own leakage rate

The single most important unknown in this whole comparison is your own leakage rate, and it's measurable: compare a small discounted test to a holdout group that doesn't see the offer, for a defined window, and calculate what share of discount-period buyers would likely have converted at full price based on the holdout's conversion rate. Stores that skip this step and simply look at 'sales were up during the discount' are measuring gross lift, not incremental lift, and will systematically overrate how well discounts perform relative to the ad-spend alternative.

Where Adsify's profit tracking fits this comparison

Because this comparison depends on accurate product cost and margin data — not just revenue — pulling those numbers correctly from Shopify matters as much as the ad platform data itself. Adsify's profit and POAS tracking uses Shopify product cost data specifically so that a ROAS-based comparison like the ones above reflects actual gross profit rather than top-line revenue, which is the same distinction that made the discount scenario in this article look worse than the revenue number alone would suggest.

A simple decision checklist

Before running a storewide discount as a growth lever, three questions: is there a low-leakage way to target the same discount (segment, timing, or offer structure) instead of storewide; is the ROAS available on incremental ad spend right now clearly below what a discount could realistically achieve net of leakage; and is there a non-profit reason (inventory clearance, list growth, press) that justifies the discount even if the pure profit math doesn't favor it. If none of those apply, the worked examples above suggest incremental ad spend at a reasonable ROAS is very often the higher-profit lever.