Skip to main content
FHAVI

Free resource

Discounts cost more than the percentage

A 15% code removes 15% from margin, not from price — and the customer who used it was often going to buy anyway. This sheet sets the rules before the campaign, including the one most stores miss: a public coupon can make your feed price wrong.

The rules

What to decide in advance

01

Per discount type

  1. Purpose: acquisition, recovery, or loyalty. A code that serves all three serves none.
  2. Depth: the maximum percentage you will ever offer on this product, computed from contribution margin, not from what competitors do.
  3. Eligibility: first order only, minimum basket, single use per customer, or restricted to a market.
  4. Stacking: whether it combines with a sale price or free shipping. Most losses come from two discounts you did not intend to combine.
  5. Expiry, enforced by the platform and not by trust.
  6. Who can issue it: which team member or app can create codes, and where they are recorded.

The feed problem

Where a coupon breaks Google

Google compares your feed price with the price a crawler sees on the landing page. If a public code is advertised on that page, the lowest obtainable price is the feed price. A mismatch arrives as a price-accuracy flag on the item, and it usually appears days after the campaign started.

02

How to stay consistent

  1. A sitewide public sale price belongs in the feed as sale_price with a sale_price_effective_date.
  2. A single-use or email-gated code does not change the public price — it does not need to be in the feed.
  3. A code displayed on the product page for everyone is a public price. Feed it, or stop displaying it there.
  4. Time the feed update before the page change, not after.

Measure

Did it work, honestly

03

The three comparisons

  1. Incremental orders against a held-out group or the previous week — not total orders during the sale.
  2. Contribution after discount, per order, against the full-price baseline.
  3. Refund rate on discounted orders; discount-driven buyers return more, and that cost arrives later.

Keep going

The rest of the library

Every checklist here is free and ungated. The research pages carry the reasoning behind them.

All resources · Research library

Discounts that stopped converting?

Send your last three promotions with margin and refund data. We will tell you which one paid for itself.

Prefer email? Write to md@fhavi.com