Five apps, one trope.

All five exist because the same insight is true: ad networks take a tax that doesn’t need to exist, and businesses can reach consumers more efficiently by going direct. InKind goes deepest in restaurants (buy credit, get bonus credit). Claim raised a $12M Series A from Bessemer in 2024 with a broader direct-rewards thesis and was acquired by Grubhub via Wonder in February 2026. Honey is a checkout extension that pulls coupon codes from across the web (with the 2024 affiliate-cookie controversy worth knowing about). Groupon is the local-deal marketplace — pay upfront, redeem at the merchant. ZKPuzzle wraps the direct-deals model in a daily-game habit + group chat with family and friends + walk-to-earn + real ownership in the network you help grow, captured at three levels (the ZKPuzzle network, the shops ecosystem, and individual stores) — like if Uber drivers held Uber shares or DoorDashers held stake in restaurants they delivered for.

Step back and the category is moving in two directions at once. The marketplace generation — Groupon since 2008, Honey since 2012, InKind since 2017 — built around a transactional moment: you pre-pay a deal, you apply a coupon at checkout, you redeem credit at a restaurant. Real utility, but the product is something you go to when you need it. Groupon’s peak-to-now arc and Honey’s 2024 affiliate-cookie scrutiny are signs of a generation maturing.

The daily-ritual generation — Claim’s 2024 Series A thesis, ZKPuzzle’s 2024 consumer-scale launch — is built around something else: a habit. You open the app every day. The deal flow is wrapped in a reason to come back. Claim got absorbed into Grubhub before the standalone version got time to compound (we’ll see what direction it takes inside the parent). ZKPuzzle is the standalone consumer-side daily-ritual play that’s still building.

Both generations have utility. The right framing isn’t “new replaces old” — it’s “different shapes for different needs.” You probably want one from each.

InKind has been running the buy-credit-upfront model in hospitality since 2017. Restaurants get cash flow earlier than their till-receipts model can support; users get bonus credit (e.g., pay $100, get $125 in dining credit). It’s a real capital-efficiency innovation — restaurants that would have gone to predatory merchant cash advances can use InKind instead.

What InKind is best at: restaurant-specific depth. The UX is built for “I’m going out to dinner” flows — venue discovery, credit application at checkout, credit-balance tracking by restaurant. If your deals-app use is heavily skewed to restaurants, InKind’s partner depth is unmatched in the direct-shops-to-user category.

Honest concession from us: for restaurants specifically, InKind has a deeper bench than ZKPuzzle does today. Where InKind has good metro coverage, the user experience for restaurants is hard to beat.

Claim is the closest competitive thesis to ours. Their 2024 Series A announcement articulated the case for direct-shops-to-user clearly: ad networks extract a tax that doesn’t need to exist, and a direct-rewards model captures more value for both shops and consumers. Bessemer led the $12M round.

In February 2026, Claim was acquired by Grubhub via Wonder. The standalone product’s outlook is TBD as it integrates into the Grubhub stack — exactly what direction the direct-shops-to-user thesis takes inside a delivery-marketplace parent is something the market will figure out over the next several quarters.

What Claim was best at: direct-rewards distribution at scale, with a broader category footprint than InKind’s restaurant focus. The thesis is structurally similar to ours, and we respect the team for arriving at it independently.

Honest concession from us: the acquisition changes the picture. We’re no longer comparing to a well-funded standalone competitor — we’re comparing to a thesis being absorbed into Grubhub, which may sharpen the standalone case for ZKPuzzle as the consumer-side direct-rewards platform that isn’t inside a delivery marketplace.

Honey was acquired by PayPal in 2020 for $4 billion. The extension applies known coupon codes at checkout — frictionless when it works. Honest concession: Honey’s checkout UX (when it’s honestly applying valid coupons) is genuinely smooth.

The asterisk: a December 2024 investigation alleged that Honey was hijacking affiliate cookies — overwriting the original creator/influencer referral with its own, often without delivering a coupon. PayPal-owned Honey has faced lawsuits as a result. We’re not piling on (the lawsuits are still resolving), but it’s a structural reason checkout extensions invite suspicion: they sit between you and the merchant, and the incentives aren’t always transparent.

Honey is also the only one of the five that is partially a scraping business — pulling coupon codes from across the web rather than having shops upload them directly. That hybrid model is part of the reason the affiliate-cookie flow exists in the first place.

Groupon owns local-deal discovery — pay $50 for $100 of restaurant credit, $25 for a $60 spa, etc. The unit economics for local merchants are real (Groupon takes a cut; merchant gets distribution). Honest concession: for restaurants, spas, kayak rentals, dance classes, and similar local-experience deals in major metros, Groupon’s breadth is hard to replicate.

The structural critique: Groupon’s discount-stacking model often trains merchants to expect customers who won’t convert to full-price regulars. The pre-pay-for-deal model is a different game from direct-shops-to-user rewards — it’s closer to a coupon clearinghouse for local experiences than a direct-rewards relationship.

Same direct-shops-to-user core. Different consumer wrapper. ZKPuzzle is a daily-game app first — users open it every morning for the daily challenge, the group chat with family and friends, and the chance at rewards from real shops. The reason this matters: engagement comes from the game habit and the social layer, not from remembering-to-check-for-a-deal. That’s how shops keep getting visibility with users without paying for an ad network to keep nudging them.

Shops, marketers, and users upload gift cards and giveaways directly through the ZKPuzzle apps (the consumer app, ZKPuzzle for Business, and ZKPuzzle Capital). Verified humans (ZK proof of human) earn them through play. No browser extension, no affiliate cookies, no checkout-flow injection, no hidden middleman taking a cut. The platform is freemium (ZKPuzzle Pro at $5.99/mo) — subscription-funded, so our incentives are aligned with keeping subscribers happy, not with selling your attention.

Four years building this (Nucleo 2022 → ZKPuzzle 2023 → ZKPuzzle 2026). The ZKPuzzle Arcade has been the highest-volume zero-knowledge application in the world after Worldcoin since Aleo mainnet launched in September 2024. Full arc on the blog; receipts on /research.

What ZKPuzzle is best at: the daily-engagement loop that creates compounding habit, plus the only social layer in the category. Group chats with family and friend groups around the daily puzzle. Walk-to-earn for stepping outside. Real ownership in the network you help grow, captured at three levels (ZKPuzzle Points = stake in the network; Shop Points = stake in the shops ecosystem; Store Points = per-store discretionary currency) — like if Uber drivers held Uber shares or DoorDashers held stake in restaurants they delivered for. ZK proof of human keeps bots out of giveaways. Quests reward referrals, content creation, gift-card uploads, app trials, surveys.

Honest concessions: if your only need is “coupon at checkout” — Honey alternatives are the right tool. If you eat out at independent restaurants in a strong InKind metro — InKind’s restaurant-specific UX is currently better than ours. If you want a discount on a specific local experience (spa, kayak rental, dance class) — Groupon’s local-deal breadth is deeper. ZKPuzzle is a different category from each of them, not a feature substitute for any one of them.