Coupon & Offer Mechanic Designer

Example output

Offer mechanic recommendation — Bramblefield roasted-garlic trial, national grocery

The goal shapes the answer more than the discount size does
Your stated goal — drive trial of a new variant without training shoppers to expect a permanent discount — actually argues against the deepest, most visible mechanic on your list. That's worth stating up front, because BOGO is often the default reach for "drive trial," and here it's likely the wrong tool for this specific goal.

Option 1: BOGO
Psychology: BOGO reads to shoppers as one of the clearest, most easily understood deals in retail — "buy one, get one free" requires no math and feels like a large, unambiguous win. It is extremely effective at driving units and trial specifically because of that clarity.
Fit to your goal: this is the strongest trial driver of the three, but it directly conflicts with your "don't train permanent-discount expectations" goal. BOGO is also the mechanic most associated with training deal-seeking behavior, because the perceived value gap between "BOGO price" and "regular price" is so large and so memorable that shoppers specifically wait for its return.
Margin impact: severe — a BOGO is functionally close to a 50% price cut on the units sold on deal, which is a heavy hit at a 38% margin. [Confirm your funding structure — brand-funded, retailer co-op, or split — before proceeding, since BOGO margin impact is the largest of the three options by a wide margin.]

Option 2: 25% off
Psychology: percent-off deals are processed by shoppers as a rate rather than a fixed reward, which tends to make them feel smaller than BOGO even when the actual discount is substantial — this is a well-established pattern in how people perceive percentage discounts versus a "free" framing, though treat it as a pattern to expect, not a guarantee for your specific shopper.
Fit to your goal: a middle option. Meaningful enough to lower trial risk on a new variant, but a smaller perceived "unlock" than BOGO, and percent-off framing is somewhat less associated with training shoppers to wait, since it reads as "a good deal" rather than "the deal I always wait for."
Margin impact: moderate — a straightforward 25% reduction in unit margin, easier to model and fund than BOGO.

Option 3: $1 off coupon
Psychology: a fixed dollar-off coupon is perceived differently depending on price point — on a $4.49 item, $1 off is roughly a 22% discount, which is psychologically similar to the 25%-off option above, but a coupon adds a small amount of friction (clipping, loading to a loyalty card, or redeeming) that percent-off shelf pricing does not. That friction slightly reduces redemption but also slightly reduces the "everyone just expects the lower price" training effect, since not every shopper redeems it.
Fit to your goal: closest match to your stated goal. The friction naturally limits how many shoppers experience it as "the normal price," and the trial-risk reduction is comparable to the 25%-off option without the same shelf-price training effect, since it doesn't change the visible shelf tag itself.
Margin impact: moderate, and more controllable — you only pay out on redeemed coupons rather than on every unit sold during a window, which is more margin-efficient than a blanket percent-off if redemption rates are typically well under 100% for this coupon type. [Confirm your typical redemption-rate assumption; do not assume 100% redemption when modeling cost.]

Recommendation
The $1 off coupon best fits your stated goal: it lowers trial risk on the new variant close to what 25% off achieves, but does so with less shelf-price training risk and more controllable margin exposure, since payout scales with actual redemption rather than every unit sold.

Shelf and coupon wording
"Try the new Roasted Garlic — $1 off your first jar." Leads with the trial invitation, not the discount, keeping the emphasis on discovery rather than price.

Want me to model the redemption-rate math against your margin to confirm the payout before you commit?