Merkl Alternatives in 2026: Distribution vs Attribution

Written by the team at Fuul, which is one of the platforms in this guide. We have tried to describe the others the way their own docs describe them, and to say plainly where they beat us. Every third party number below links to its public source at the end.

‍

Key takeaways

Merkl is the largest distribution engine in the category, with $1.6B distributed across 250+ companies and 60+ chains. If your question is how to pay eligible wallets across many chains without building the rail yourself, it is usually the right answer, and nothing on this list matches its reach.

Most teams searching for a Merkl alternative are looking for something Merkl does not sell. They want attribution: a record of which person caused which activity, and a commission agreement that pays that person for it. Distribution and attribution are separate products with separate data models.

The real options in 2026 are Fuul, Turtle, Galxe and building it yourself. Most programs past their first season end up running two of them at once.

‍

The options at a glance

And the capability matrix, which is where most evaluations actually get decided:

‍

Distribution and attribution are two different products

A distribution platform starts from the state of a blockchain. At the end of a period it reads every position in a pool, a vault or a lending market, applies a formula, and produces a list of addresses and amounts. The input is the chain. The user is whoever happened to be holding.

An attribution platform starts from an event and a person. Somebody clicked a link, entered a code, opened an account, made a deposit, finished a checkout. The job is to record who caused it, resolve that person to an identity that survives across wallets and sessions, and hold the commission agreement that says what they are owed.

Both end in a payout, which is why they get compared. The difference shows up in the questions each one can answer.

A distribution engine can tell you that 0xab… held $40,000 of USDC in a vault for eleven days and is owed 312 tokens. That is a real computation and it is harder than it looks. What it cannot tell you is that the deposit arrived through an affiliate who has brought 84 other depositors this quarter, sits on a 12% tier, has $9,400 unpaid, and needs a W-8BEN on file before the next claim goes out.

Neither is a better version of the other. They are answers to different questions, and the confusion is expensive in a specific way. A team buys a distribution engine, launches a referral season on top of it, and within six weeks a spreadsheet is doing the attribution. The spreadsheet becomes the product. Nobody planned that, and nobody can audit it in December when finance asks who got paid what.

Settle one thing before you shop. Does anybody need to get paid for bringing somebody else? If no, you are buying distribution and the list is short. If yes, that is a second system, and it does not come free with the first one.

‍

How Merkl works

Merkl is the baseline here, so it gets described properly rather than as a foil.

A campaign creator deposits the reward token, defines the eligibility rules, and Merkl's offchain engine computes who earned what. The result is published as a Merkle root onchain, and users claim against it. Claims can also be delegated, so a protocol can push rewards to users who never visit a claim page.

Campaign types cover concentrated liquidity, lending markets, plain holding, and airdrops from a JSON file. Distribution can be variable, fixed, capped, or set to a target APR.

Two pieces of engineering explain why large protocols pick Merkl and then stay.

The first is how positions are measured. Merkl calculates a position as an integral over the campaign window rather than sampling it at intervals. A user who deposits an hour before a snapshot and withdraws an hour after it earns an hour of rewards, not a full period. Sampled systems get farmed on exactly this, and most teams discover it the hard way.

The second is reward forwarding. Their forwarders follow beneficial ownership through nested contracts, so a user who deposits the incentivized asset into another protocol keeps earning. Without that, half of DeFi composability quietly breaks your campaign.

Eligibility filters are granular: OFAC screening, optional World ID, allow and deny lists, minimum balance, holding duration, and a new users only filter.

Target total APR is in production. It runs live on Aave, where USDm lenders earn a guaranteed 5% total APR paid in $MEGA, with rewards auto adjusting as the native yield moves. The protocol sets the floor and Merkl tops up the gap, paying nothing in the months native yield covers it. This matters if you are comparing roadmaps, because several platforms describe target APR as something they are building. Merkl already runs it.

Pricing is a take rate on what you distribute, and it comes out of the campaign budget rather than being added on top. It steps down: 3% on the first $1M, 2.25% from $1M to $2.5M, 1.75% from $2.5M to $5M, and 1.5% above that. Airdrops distributed by JSON are 0.5%. Points programs are priced separately at $40 per week per source. There is a 25% discount for committing early above $500k.

Scale, for context: $1.6B distributed, 250+ companies, more than 60 chains, 4M+ wallets, and over 200,000 monthly active users on the Merkl app. Their first non EVM chain was Stellar in May 2026, with Flare arriving the same month. Clients include Aave, Morpho, Ethena, Uniswap, Optimism, Sky, Circle, PayPal, Coinbase, Kraken and Robinhood. Funding is a single $5M seed led by a16z in 2021, raised as Angle Labs. That last one gets misreported constantly, because a much larger figure circulates in startup databases and belongs to an unrelated American health insurer with a similar name.

What Merkl says it does not do

Their docs are unusually direct about scope, and the boundaries are deliberate rather than accidental.

There is no general offchain event API. The calculation is described in onchain terms, so a Stripe checkout or an action inside a fintech app has no path in.

There are no tax forms and no KYC, by explicit design. Their pages say Merkl is pseudonymous and does not know its users. If you need a W-9 on file before a claim, that is a different system.

They never mint, hold or distribute points. Points are indexed; the canonical balance, the leaderboard and the airdrop logic stay with you.

Social actions are out of scope by their own statement. No X follows, no Discord joins.

On referrals, be careful with what you read elsewhere. Merkl does have referral attribution, single level, available onchain and through their API. Anybody claiming otherwise has not read their docs. The difference is structural. Their referral boost redistributes the campaign budget rather than paying a commission on top of it, and there is no affiliate portal, no payment terms, no approval queue and no separate accounting. It is a configuration flag on a campaign. If what you need is a referral multiplier inside a liquidity campaign, it works. If what you need is to run an affiliate business with sixty people who invoice you monthly, it is not that.

‍

The alternatives in 2026

Fuul

Fuul is attribution infrastructure for incentive programs. Its unit of measurement is a person and the chain of people behind them, rather than a position.

An event arrives, onchain or through an API. Fuul resolves it to a user, walks the referral graph to find who brought that user, and applies the commission terms in effect when the referral happened rather than the terms in effect today. That last part sounds like a detail until you change your rates.

Around it sit the pieces an affiliate program needs and a distribution engine does not have: a portal under your brand, multi level payouts with agencies earning a spread on sub affiliates, per affiliate monthly invoicing, tax forms (W-9, W-8BEN, W-8BEN-E) with a review queue that blocks claims until approved, and reward rates that vary per audience segment by balance, tenure, prior activity, chain or a custom query.

Because events can arrive from anywhere, users do not have to be onchain. belo, a neobank with three million users, ran a referral program through Fuul without changing their app: 8,499 confirmed attributions, a 99.6% claim rate, zero fraud cases. Nado runs invite codes the same way, with $6.07B in referred volume across 6,954 affiliates. Concrete migrated a whole program off a vendor that was shutting down, 1,048,483 addresses and 530,858,951 points.

Pricing is a flat monthly fee tiered by program complexity, the opposite trade to a take rate. Below roughly $1M distributed per year, Merkl comes out cheaper.

Where Fuul does not compete: chain coverage (EVM and SVM, nowhere near 60), concentrated liquidity campaigns, and target total APR, which Merkl runs in production today and we do not yet ship at all.

Turtle

Turtle is not an incentive platform and treating it as one leads to a bad decision in both directions.

It is a three sided liquidity marketplace. On one side, protocols that want deposits. On another, distributors: wallets, exchanges and KOLs who route users. On the third, the LPs. Every deposit is attributed onchain to the distributor that brought it, and Turtle charges a fee on the TVL attributed through their network. They are paid for bringing the money rather than for moving it, which makes it a performance fee on capital acquired rather than a platform fee on volume distributed.

The reason it shows up in this search is that it comes out of the same budget line, and its distributor side is the same idea as an affiliate program. The difference is ownership. Turtle runs that network as its own asset and rents you access to it. The distributor's relationship is with Turtle, the attribution data lives with Turtle, and when your campaign ends the distributor stays there along with everything they learned about routing users to protocols like yours.

That is a real trade rather than a trick, and for some teams it is the right one. If you are launching with no wallet integrations, no KOL relationships and no list, renting a network of more than 360,000 connected wallets is much faster than building one. One contract instead of fifty conversations.

They have raised $11.7M across two rounds and routed $5.5B of cumulative liquidity. Their self serve product, Streams, is a TVL campaign at 1.5%, included free in a Pro tier around $10k. They also publish good research, including a March 2026 piece arguing that 70% of incentivized TVL leaves within 30 days.

Galxe

Galxe sits in a different category and competes for a different budget. It is a quest and social campaign platform with its own L1, and the campaigns it runs are the ones a distribution engine explicitly refuses: follow an account, join a Discord, complete a task, mint a credential, climb a leaderboard built from actions rather than balances.

That makes it a marketing tool rather than a treasury tool, and it is genuinely the strongest option for the top of the funnel. If the problem is that nobody has heard of you, paying people to bridge $500 into a pool does not fix it, and a quest campaign in front of a large existing audience might. Galxe has that audience and the campaign tooling to put something in front of it quickly.

Pricing is SaaS. Their Business+ tiers run $999 per month for Essentials and $1,699 for Growth on annual billing, or $1,166 and $1,999 on quarterly, with Enterprise quoted custom. There is no monthly option. That is a different shape from both a take rate and a program platform fee, and for a marketing team with a quarterly campaign calendar it is an easy line to approve.

Where it stops is where treasury scale incentives begin. Quest completions are cheap to fake, the reward is usually a badge or a small fixed allocation rather than a rate on capital, and the attribution ends at the task. Knowing that 40,000 wallets completed a quest tells you nothing about which of them deposited, stayed, or brought somebody else. Teams that run Galxe seriously tend to run it as the awareness layer and put something else underneath it to measure what happened next.

Building it in house

The honest version of this option starts by admitting it is often correct. One chain, one reward type, one season, a formula that fits in a page, and an engineer with two weeks. Ship it. A spreadsheet and a Discord bot have launched more successful season ones than any vendor on this list.

The cost arrives later, and it recurs. It shows up with the second chain, the second reward type, the first sybil farmer, and the first user who finds a way to claim twice. The modules stack up quickly: attribution, anti fraud, multi chain payouts, a claim contract someone has to audit, invoicing, tax forms, and a support queue for the affiliate who says his commission is wrong. Each one is buildable. The problem is that the team that built them is now the team that maintains them, permanently, and that draw comes out of the engineering hours that were supposed to go into the actual product.

The moment to watch for is when the incentive program stops being a project and becomes a system, because a system has an on call rotation and a project does not.

Two questions that usually settle it. Will this program still be running in eighteen months? And when the first farmer shows up, does anybody on your team want to own the response? If both answers are yes, build it. Most teams find one of them is no.

Also considered

Four more names come up in the same evaluation often enough to be worth placing, even though none of them is a like for like swap for Merkl.

Royco builds incentivized action markets. Instead of a protocol setting a rate and hoping it clears, incentive providers and action providers post offers and counter offers until they agree on a price for a specific onchain action, and the transaction and the incentive settle atomically. It is a pricing mechanism rather than a distribution engine, which is a genuinely different idea and the reason Merkl publishes a comparison page against them. Live on Ethereum, Arbitrum and Base.

Absinthe is points issuance and management. It ingests onchain events and some offchain engagement, runs no code, and gives you a dashboard and a user facing portal. If your specific complaint about Merkl is that they will index your points but never own the ledger, Absinthe is a direct answer to that complaint in a way nothing else on this list is.

Layer3 and Zealy sit in Galxe's column. Zealy is self serve community gamification, closest to the Discord grind phase. Layer3 is more curated, with a higher entry barrier and quests that tend to go deeper onchain. Both compete with Galxe for a marketing budget, and neither one changes the attribution problem described above.

‍

Which should you choose

You are distributing rewards for onchain positions across many chains, and nobody is going to ask who brought the depositor. Merkl. This is what it was built for, the chain coverage is not close, and the position math is better than what you will write. Below roughly $1M distributed per year it is also the cheapest option here.

Growth runs through people: referrals, affiliates, invite codes, agencies, or a community you want to formalize and pay. Fuul. The commission structure, the portal, the invoicing and the tax layer are the product rather than a flag on a campaign, and the attribution follows the person rather than the position.

Your users have an account before they have a wallet. Fuul, and the reason is narrow. Payouts can be created against an email and resolved to an address later, and events can arrive from a fintech backend rather than from a chain. This is the case that has no answer in a purely onchain system, which is why fintechs end up here.

You are launching, you need liquidity within sixty days, and you have nothing to distribute through. Turtle. Renting a network beats building one when you have nothing to rent from yourself. Come back to the ownership question at renewal rather than at signing.

The problem is awareness, not capital. Galxe. Quests, social actions and a large existing audience, on a SaaS fee that does not scale with your treasury. Then measure what happens after the quest somewhere else.

‍

Using more than one

Most programs past their first season run two of these, and the combinations are fairly predictable.

The common one is a distribution engine for liquidity depth with an attribution layer on top for the people bringing that liquidity. Merkl computes what each position earns; the referral and affiliate side runs separately, with its own terms, its own portal and its own accounting. The two touch at the point where a payout is created, and nowhere else. Teams that try to force both jobs into one tool usually end up doing the second one in a spreadsheet.

The sequenced one is renting distribution while building your own. Turtle brings the first liquidity in while you stand up an affiliate program underneath, so that at renewal you are negotiating rather than depending. We have clients doing exactly that, and it is a better plan than either half alone.

The funnel one is Galxe at the top and something with real attribution below it. A quest campaign is a fine way to get 40,000 people to show up. It is a poor way to find out which 400 of them mattered.

What does not work is running two systems that both think they own the payout. Decide early which one is the record of what was paid, to whom and why, because that is the question you will be answering six months later when someone from finance asks.

‍

Frequently asked questions

Does Merkl support referrals?

Yes, single level, onchain and through their API. The structural difference is covered above: the boost redistributes the campaign budget rather than paying a commission on top, and there is no portal, no payment terms, no approval queue and no separate accounting.

Is Merkl expensive?

It depends entirely on volume, and the crossover is worth calculating before you sign anything. The take rate steps from 3% down to 1.5%, taken out of the campaign budget. Below roughly $1M distributed per year, that is cheaper than any flat monthly fee in this category. Above it, a percentage of a growing number grows with it. A program distributing $10M pays about $182,500, an effective rate of 1.8% once every step is applied.

Can I run a points program on Merkl?

You can index one. Their docs are explicit that they never mint, hold or distribute points, so the canonical balance, the leaderboard and the airdrop logic stay on your side. Indexing is priced separately at $40 per week per source. If you want the platform to own the points ledger itself, that is a different product.

What about users who do not have a wallet yet?

There is no path for them in a purely onchain system, and Merkl says so plainly by describing itself as pseudonymous. Fuul creates payouts against an email or an internal account identifier and resolves them to an address whenever the user gets one. For a fintech or a neobank, this is usually the whole reason the evaluation started.

Can I move a running program without resetting it?

Yes, and it is the most common way programs change platforms. Concrete moved 1,048,483 addresses and 530,858,951 points off a vendor that was shutting down, with the history intact. Ask any platform you are evaluating for the import path before you commit, because the ones that do not have one will tell you your users can just start over.

Is target APR available anywhere other than Merkl?

Not really, today. Merkl shipped it and runs it in production, including a live Aave campaign where USDm lenders earn a guaranteed 5% total APR in $MEGA with the rewards auto adjusting as native yield moves. Fuul has a version in development that adds a referral boost to the target, so a referred deposit lifts the tier. It is not shipped, and this guide will say so until it is.

Do I need anti sybil defense beyond OFAC screening?

Screening and fraud detection solve different problems. OFAC lists, optional World ID and duration gates keep sanctioned or obviously ineligible addresses out, which is compliance. None of them catch one person operating forty wallets that are all individually compliant, which is where incentive budgets actually leak. If your program pays per user rather than per dollar of TVL, the second problem is the one that costs you money.

‍

Where to go from here

If the answer to the question at the top was no, nobody needs to get paid for bringing somebody else, stop reading and go configure a Merkl campaign. It will be cheaper and faster than anything else here.

If the answer was yes, the thing to work out next is what your attribution has to survive: rate changes mid program, affiliates who invoice you, users who arrive before they have a wallet, agencies earning a spread on the people they recruit. Fuul was built for that specific list. If you would rather see it running, book a demo.

‍

Sources

Every third party figure in this guide comes from a public page. Checked 2026-08-18.

Merkl fee model and the degressive take rate: docs.merkl.xyz/distribute-with-merkl/fee-model

Merkl scale ($1.6B, 250+ companies, 60+ chains, 4M+ wallets): merkl.xyz

Merkl distribution types and campaign mechanics: docs.merkl.xyz/merkl-mechanisms/distributions

Stellar and Flare as the first non EVM chains, May 2026: Merkl monthly recap, May 2026

Target total APR and the Aave USDm / $MEGA campaign: Merkl monthly recap, April 2026

Angle Labs $5M seed led by a16z, September 2021: a16z announcement

Turtle funding, $6.2M seed plus $5.5M, $11.7M total: GlobeNewswire

Galxe Business+ tiers and billing: help.galxe.com

Royco incentivized action markets: docs.royco.org

Absinthe points issuance: absinthe.network

‍

Last updated 2026-08-18. Fuul publishes this guide and appears in it. If anything here is out of date or unfair to Merkl, Turtle, Galxe or anyone else named, tell us and we will correct it.