Pricing

Discount code strategy that doesn't leak revenue

How to use sponsor codes, partner codes, past-attendee codes, and influencer codes without training your audience to wait for discounts.

7 min read min read · Published March 23, 2026

A conference organizer I worked with was proud of her discount code program. She had 47 active codes across her sponsors, partners, influencers, and past attendees. 62% of paid registrations came in through a discount code. Her average ticket price was 31% below the advertised price. What she couldn't figure out was why her margin per registration kept dropping year over year while advertised prices kept climbing. The discount program had quietly become the default price of her event.

Discount codes are a sharp instrument. Used well, they attribute sponsor value, reward loyal attendees, and shift fence-sitters off the fence. Used badly, they train your audience that the advertised price is a suggestion and your revenue becomes whatever the discount stack produces. This guide covers what works and the traps to avoid.

1. The four code types — and what each is for

Each discount code category has a different purpose and should be designed differently.

Sponsor codes (15-20% off)

  • Purpose: attribution and sponsor value delivery.
  • Naming: sponsor-branded — ACME20, not SAVE20.
  • Volume: unlimited per sponsor (it's their outreach).
  • Use: sponsor's email list, LinkedIn posts, their website.
  • Tracking: every registration through the code attributed to that sponsor for the renewal conversation.

Partner codes (15-20% off)

  • Purpose: reciprocity with industry partners (media, associations, adjacent conferences).
  • Naming: partner-branded — TECHCRUNCH15.
  • Volume: unlimited or capped depending on agreement.
  • Use: partner's channels.
  • Tracking: same as sponsor.

Past-attendee codes (20-25% off)

  • Purpose: loyalty recognition and return-rate driver.
  • Naming: cohort-branded — ALUMNI25 or year-based.
  • Volume: limited time window (first 4 weeks of registration only).
  • Use: direct to past-attendee email list.
  • Tracking: match against prior registration list; expire at deadline.

Influencer codes (variable, typically 10-15%)

  • Purpose: reach expansion via practitioner/creator audiences.
  • Naming: influencer-branded — JENKINS10.
  • Volume: unlimited within agreement period.
  • Use: influencer's channels, often in exchange for ticket + honorarium.
  • Tracking: driver of whether to work with that influencer next year.

2. Stacking rules (the single most important decision)

Disable stacking. This is the rule that separates a profitable discount program from a leaky one.

If an attendee has a 20% sponsor code and is registering during early-bird (20% off), the applied discount is 20%, not 40%. Either the tier discount or the code applies, whichever is greater. Communicate this clearly on the registration page.

Stackable discounts are how 47-code programs end up averaging 31% below list price. Non-stackable discounts cap the exposure at whatever the largest single discount is — usually 20-25%.

3. Code naming conventions

Every code should be:

  • Branded to the source. ACME20 ties the code to Acme Corp for attribution. SAVE20 is generic and leaks.
  • Short but not guessable. 6-10 characters. Not SPONSOR20 (guessable). Not X4GJ29KD (unshareable).
  • Documented with terms. "ACME20 — 20% off, valid through {date}, Acme sponsors only, non-stackable."
  • Expiring automatically. All codes should have a sunset date. An active code with no expiry is an invitation to be shared on discount-aggregator sites 18 months from now.

Code naming matters because codes leak. SAVE20 that appears on RetailMeNot gets used by people with no connection to the program and drives 30-50% of discount spend to audiences who would have bought anyway. Branded codes still leak, but the leakage is traceable.

4. Attribution: the dashboard that drives renewals

The value of discount codes is in the attribution, not the discount. A per-code attribution dashboard should show, for each code:

  • Registrations driven
  • Revenue (gross, after discount)
  • Average ticket price through the code
  • Quality of registrant (ICP fit, seniority, company size)
  • Renewal intent (did they attend last year?)

Share the attribution dashboard with sponsors at renewal time. A sponsor who drove 80 registrations through ACME20 renews at 85-90%. A sponsor who drove zero renews at 20-30%. The data makes the renewal conversation concrete instead of subjective.

Modern event registration platforms export this data natively; if yours doesn't, a simple spreadsheet with registration-code joins works.

5. The traps

  1. The "open code" on the homepage. EARLY15 in the site footer trains every visitor that 15% off is the default. Never put a discount code in site-wide navigation or footer. Codes belong in distribution channels, not on the landing page.
  2. Code proliferation. 30+ active codes is a sign the program is out of control. Cap the number of active codes per quarter and audit them monthly.
  3. Influencer overpayment. Large influencer codes to people who would have promoted anyway burn margin. Start with a 10% code; increase only if the data supports it.
  4. Too-long validity windows. Codes active for 6 months get shared on external sites. 30-60 day windows are the practical maximum.
  5. Refunds through the discount. An attendee paid $800 through a $200 discount code; they request a refund; you refund $800 but attribute the $200 loss to the wrong place. Track refunds against codes.
  6. VIP/internal codes leaking externally. The 100% comp code INTERNAL100 has a way of being shared. Better: manually issue single-use codes for internal comps, with attribution to the person who authorized each one.

6. When to run site-wide promos

Sitewide codes (unbranded, open to anyone) should be rare. They work for:

  • Flash 48-72 hour promos tied to a specific moment (a Cyber Monday push, a specific industry event milestone).
  • Launch announcements (first 48 hours after registration opens, capped at 20% of total seats).
  • Final-week seat fill if genuinely short of target (better to run a time-boxed promo than discount to everyone).

Outside these moments, run targeted codes only. Constant sitewide promos train the audience to wait.

Discount program checklist

  • [ ] Code naming convention documented (source-branded)
  • [ ] Stacking disabled on the registration platform
  • [ ] Expiry date set on every active code
  • [ ] Active code count capped (<15 at any time for most programs)
  • [ ] Attribution dashboard in place before any sponsor codes issue
  • [ ] Refund tracking against codes
  • [ ] Site-wide codes limited to flash windows
  • [ ] Monthly audit of active codes
  • [ ] Code terms published internally for sales/support reference

FAQ

What discount percentage should a sponsor code offer? 15-20%. Below 10% feels insubstantial and sponsors won't promote it. Above 25% erodes margin on registrations you would have won at full price.

Should past attendees get the deepest discount? Yes — 20-25% for a short window (first 4 weeks of registration). It converts returners early, which frees your later marketing spend for acquiring new attendees.

How many codes is too many? For most B2B events, more than 15 active codes at once signals program overreach. Audit monthly; expire inactive ones.

What about 100%-off comp codes for VIPs? Issue manually, single-use, named. Never a generic comp code. A comp code VIP100 will end up on Reddit.

Do I need to disclose that codes exist publicly? No — but if asked, have a consistent honest answer. "We offer discount codes through sponsors, partners, and past-attendee channels. They are not publicly posted."

Related guides


Ready to run your next event? Create your event on WebMobi →

Related guides

Run your next event on WebMobi

Build the registration page, app, and analytics in under an hour. No credit card required.

Create your event →