Coupons have a reputation problem, and they earned it honestly: every market has a business whose "sale" never ends, whose customers have learned that full price is for suckers, and whose brand now means "wait for the code." That's not what discounts do — it's what unstructured discounts do.

Used with structure, a coupon is one of the few marketing tools that converts on its own: it gives a willing customer permission to act now. The whole craft is making sure "now" never quietly becomes "always." We covered the single-campaign version in the holiday sprint; this is the year-round strategy — codes, windows, partners, and guardrails.

The one-paragraph psychology

A discount changes behavior through urgency, not price. Most redeemers were already interested — the coupon collapses "someday" into "this week." Cheapening happens at exactly one point: when customers learn the discount is predictable. Two or three structured occasions a year read as generosity; a monthly pattern reads as your real price list, and from that day forward you've raised your sticker price and lowered your actual one. Everything below is machinery for staying on the right side of that line. (The honest-pricing foundation is in pricing with confidence — a coupon can't fix a price nobody believes.)

Code strategy: the code is copy

A coupon code is the shortest piece of marketing you'll ever write, and customers read it. SAVE10 says "we discount things." The alternatives each say something better:

Two standing rules regardless of type: codes never stack (stacking turns pricing into a puzzle game and your margin into the prize), and every code is wired to a real offer through checkout so redemption is one paste, not a negotiation.

Expiry windows: match the deadline to the deliberation

Urgency is a window-width problem. Too long and there's no reason to act (a 60-day code is a bookmark); too short and you expire before your customer's natural decision cycle completes (a 48-hour code on a $2,000 service punishes exactly the deliberation the purchase deserves). The rule: size the window to how long this purchase takes to decide, plus a little.

• Impulse-priced products and bookings — 3 to 7 days. The decision is one conversation at dinner.
• Considered services — 2 to 4 weeks, which covers a quote conversation, a partner consultation, and a payday.
• And then keep it. The expired code that still works teaches the most expensive lesson in this post: that none of your deadlines are real. Honoring expiry — politely, with a "that one's ended, but here's what's current" — is what makes every future window do its job. Deadline integrity compounds across events, launches, and campaigns alike.

Bulk and unique codes: attribution, control, revocation

The shared-code era is over for anything that leaves your hands. When a code goes to a partner's audience, a deal site, or an individual customer, generate it unique to that channel or person, in bulk where needed. Three reasons, each worth it alone:

• Attribution you can trust. Each partner's code is its own ledger line — YOGA-WITH-JEN's twelve redemptions settle the "did that collaboration work" question with arithmetic, the same way tracking links settle it for clicks. Pay partner commissions from code counts, not screenshots.
• Containment. A shared code posted to a coupon-aggregator site is a brand-wide markdown you didn't plan. Per-customer unique codes (the win-back sweetener, the referral thank-you) are non-viral by construction — one use, one person, done.
• Surgical shutdown. When a partnership ends or a code leaks, you revoke that code. Everything else keeps running.

Cap usage on everything — per-customer (once), and total (the first 50 redemptions) where supply or margin demands it. A cap is also honest urgency's second axis: "until Friday or the first 50" gives both deadline and scarcity something true to point at.

The brand guardrails

• A floor you never cross. Decide the deepest discount your brand ever shows — say 25% — and hold it everywhere, including partners. Depth is what registers as desperation; frequency is what registers as policy. Stay shallow and occasional.
• Flagships stay full price. Your signature service and newest release are what the brand means — discount the adjacent things (the add-on, the starter package, last season) and let the flagship's price tell its story uninterrupted.
• The full-price majority rule. If most of your revenue isn't at full price, you don't have a coupon program — you have a mispriced catalog wearing one. That's a pricing fix, not more codes.
• Sweeten before you slash. A bundled extra or gift-card bonus often converts as well as a percentage and protects the price architecture entirely — the same swap the holiday playbook leans on.

Measure: redemption count is not the metric

A hundred redemptions could be a hundred new customers — or a hundred regulars who'd have paid full price, now subsidized. Read campaign performance with two splits:

• New names vs. known names. A code that mostly recruits is an acquisition tool earning its discount; one that mostly rewards regulars is a margin leak (or, reframed deliberately, a loyalty perk — but choose that on purpose).
• Margin per redemption, not revenue. The discount, any partner commission, and the order size together decide whether the code made money. A 20% code that lifts average order value 30% is a winner; the spreadsheet knows, the redemption counter doesn't.

Then turn the results into follow-up: code-recruited first-timers enter the welcome journey at full-price footing — the coupon got them in the door; the relationship is what was actually on sale.

Frequently asked questions

Codes, expiry, caps, checkout, and the redemption numbers all live in one Faster workspace, next to the campaigns that carry them. Name your next code like it means something, give it a window you'll honor, and let full price stay the price.