New referal system

With the arrival of Paraswap V2, referral was introduced in a raw way into the concept.
It is now time to anchor it in a permanent way in Paraswap. Many users, anxious to have a protocol as close as possible to the economic and social realities of the DEFI, have therefore started discussions on the discord about the existing “Referral Program”.

It is important to understand that the Referral Program now offers the referee and the referer the possibility to accumulate 2 different boosts per Epoch.

  • One at 10% for volumes achieved as a referee, the second at 20% as a referrer of an address that has done volume on Paraswap. This represents a significant increase in Paraboost and therefore in its share of the protocol’s revenues.

It is therefore necessary to identify the critical points, those that should not be in order to obtain a totally fair and attractive system.

A series of inconsistencies and nonsense emerge as follows:

  • The referral link must be used for each trade for the volume of the referree to be taken into account
  • It makes no difference if we refer 1 or 100 people
  • It makes no difference if our referrals make $1 or $100m in volume
  • An address already using Paraswap can use one link and get a boost
  • An address can use several links to boost several people
  • Any address, even if it has never traded on Paraswap, can offer its sponsorship and therefore a boost
  • A part of the positive slippage is given to the sponsor 25% and to the sponsored person 25%.

Before we start, it is important to understand the new dynamics of Paraswap for the Stakers.
The more money the protocol makes, the more rewards the stakers will receive. It is therefore in everyone’s interest to work towards creating the most efficient protocol possible in terms of return.
At the same time, it is in everyone’s interest to design metrics around the “Paraboost” that give all stakers the rewards they deserve.

This being said, it is clear that the current referral system is no longer adapted. It favours the players, is too easy to access and does not allow to value the people using their network to bring new users.

One more reminder:
+Users=+volume=+fees+of visibility+of partners+of listings+of grants=+of revenue=+of rewards at each epoch

Sponsorship is a key element in the marketing of a company. It is therefore necessary to rethink it as a whole and in a serious manner.

Let’s find a consensus in this topic and come up with a proposal.
The new measures only affect 22 referee addresses (epoch 2) with a total referral volume of $7.7m
And 19 sponsor addresses for a total of $5.4m
It is therefore important to review the system “quickly” in order to impact as few addresses as possible. The longer we delay releasing a new version, the more users will be impacted.

Here are the main points for discussion and reflection

1) End the obligation to use the referral link every time.

We need a simple referral system. Ban the obligation to use the link every time you connect and favour a link between the sponsor’s address and that of the sponsored person.
When a sponsored friend clicks on the link of his sponsor and connects with his address, this address will be taken into account for the affiliation. From then on, he will have his sponsor linked to his address.
At each connection on Paraswap with this same address his volumes will be counted for his sponsor.

2) Blocking the possibility of having several sponsors from one address

Thanks to measure number 1, each address will be the godson of a unique sponsor. This will prevent abuse to boost several potential sponsors.

3) A dynamic boost according to the volume of referrals

A table of volume metrics should be designed

4) Return the positive slippage to the protocol

With Social Escrow, the user has a direct interest in offering his sponsorship to other users. He is rewarded with a boost that will increase his share of the protocol’s revenue.
This being said, the old system of sharing a part of the positive slippage is no longer valid.
All the protocol income should be returned to the common pot, it is therefore necessary to remove from the godchildren and the sponsor the 50% of the positive slippage provided for in the vote

5) Advantage for the godchild

The referree will get a unique boost for 1 to 3 epochs. This way, the new user is rewarded as well

Here is a basis for reflection. Nothing is exhaustive, nothing is set in stone. By providing some context and material, we hope that there will be many discussions and that this will help to better understand the possible “future system” of sponsorship

4 Likes

Having an easier link between the referee and referrer is definitely a big step in the right direction, I like your idea of making it more volume-based and having a stronger referrer-referee relationship, that way they both benefit from each other!

I wanted to add my own two cents as well: Something else that is missing from the current referral system is an instant understanding of the benefits of using the system. Boosts are a nice benefit that stakers can clearly understand , but brand-new users of ParaSwap might be confused. This is why I’d like to throw an idea out: What if we introduce a Pro mode, where if the total referred volume hits an objective people unlock benefits?

Here’s what I’m thinking

  • Referrer shares their link far and wide, they receive the usual benefits (boosts, part of revenue, etc).
  • Once they hit a minimum level of traded volume/fees/any other metric , they unlock a Pot of Rewards with a set budget
  • With the unlock of the rewards, all addresses unlock a new special perk: It can be 100% refunds on the base fee, it can be sePSP1, etc.
  • New and better rewards are earned as bigger targets are hit.

Of course, there will need to be some anti-sibyl and abuse measures in place, but I think that the target volume should stop the system from being easily abusable. What do you think?

2 Likes

I love your idea but in my opinion, we should stop giving out carrots all the time.

The referer and referee already receive boosts that increase their share of the income.
It goes without saying that today some may find the system unprofitable (well 125% apr doesn’t seem unprofitable to me but let’s say common belief thinks like that). But the day paraswap will be used on a regular basis, this carrot will be too much for me.

I’m all for attracting thinking people. An informed and educated population of DEFI users.
The more carrots you put in, the more interested and uninteresting your population becomes (example of the lowering of gas refunds, a heresy that allowed some profiteers to prosper).

In short, the more carrots you put in, the more individualistic your users are.

If we work to make the system reliable, fair and economically interesting, then we will have a close-knit and deeply educated community.

3 Likes

I agree that the referal system must be rework.

  • The referral link must be used for each trade for the volume of the referral to be taken into account
  • It makes no difference if we refer 1 or 100 people
  • It makes no difference if our referrals make $1 or $100m in volume
  • An address already using Paraswap can use one link and get a boost
  • An address can use several links to boost several people
  • Any address, even if it has never traded on Paraswap, can offer its sponsorship and therefore a boost

I agree with this non sense points.

Point 1 & 2
To me it’s the biggest pain point for the promoting referral link. A wallet using a referral link should be link to the referer for ever. In that way it will bring more people to share they link (like influencers aka dydx, CEX, etc.). For that link should be easy aka stakeDAO referrals system (where the referral link could be something like https://app.paraswap.io/r/twitterAccount)

Point 3, 4 & 5
In my point of view, I also agree that only one benefits should be taking into account: positive slippage part OR a boost. I like the idea of the dynamic boost according to the volume but it’s also open an easy sibyl attack (You get your 40% of boost for trading volumes on your main account, you use another wallet with your referral link and do another X% of bonus). That why I prefer the positive slippage bonus.

If we work to make the system reliable, fair and economically interesting, then we will have a close-knit and deeply educated community.

100% agree with that.

I also don’t think we need to provide more budget and carrot on it as @stikers said, but make it user friendly and easy to use/follow.

But to have a reliable system we need to work on an anti-sibyl solution

3 Likes

Hello to all,

Thank you @stikers for starting this topic on the forum!

I agree with everyone on the flaws presented in the current referral system.
_


Point 1 & 2 - Use of referral link (UI/UX/consequences)

For the ease of use of the referral system, I agree that a link system like the one presented by @Xutyr
would already be an improvement.
If we push the thought a little further (and with the idea that a ParaSwap account would arrive soon), could we not envisage that this link be used to link two wallet in the UI.
Via this link, the referee chooses his referrer. It is linked to him and him alone. However, he can change his referrer when he wants to by using another link (it remains to be seen under what conditions this change would be made).
Thus the link is used only once and the accounts or wallets are linked, all the volumes of the referee will be associated with the wallet / account of the referrer he chooses, until the referee removes this link or replaces it with another referrer.

I wouldn’t say that two wallets (referrer/referee) have to be linked forever (I think that’s easy to dodge) but at least that they have to be linked for an epoch (or a defined number of epochs).
_


Point 3 & 4 - boost & incentives

I am also of the opinion that the referrer / referee system should only benefit from the Paraboost advantage.
This would simplify the system where, any action by the social escrow gets a larger share of the protocol revenue via the Paraboost score (I don’t take GRP into account). So we would have a positive slippage / fees redirected only to the stakers, and distributed according to the Paraboost share.
_

Maybe I want to oversimplify and add a “Pro mode” with extra incentives would be a useful perk to motivate referrers.
I just think that being social escrow centric would be to focus on increase one’s Paraboost score and that we should check if improving / adapting the Paraboost factor is sufficient to incentivize correctly the users (Referrer/referee in this case). If not, or if the solution we have can be abused and we don’t find a sybil-resistant one, I would consider addind an extra layer of incentives.
_

This brings us to the tricky part, the fairness of the boost referral in all this. Indeed, it would be necessary to imagine an anti-sybil system which would allow an encouragement to the participation by big referrers.
_

As stated above by @Xutyr, current boost provided by referee to referrer could make the system easily abused.
One could imagine that a small percentage of the volume realized (10% or so) by his referree would be taken into account as referrer’s own volume and thus taken into account in his paraboost score.
This would require a very large volume or many referees to really impact the referrer’s paraboost score.
The boost given to the referrer/referee is progressive and depends on the volume handled by the referee.
Here I focus mainly on the incentive made to the referrer because for me it is he who makes the activity related to the social escrow. The referee benefits from PoaraSwap rates and the boost which for me is sufficient.

The negative point though, the big bag potentially big referrer already easily reach their volume related boost so the interest may be negligible and it may need more thought on the volume boost in parallel. Referrer needs a fair incentive to onboard, encourage referees to use ParaSwap.
Difficult balance to find between rewarding the referrer and not unbalancing the social escrow / paraboost system.
_


For now, I would be more in favor of seeing what can be done by rebalancing incentives rather than creating new ones.
But maybe I’m underestimating the impact on the volume and usage of the dapp of referrers engaged in the protocol and an additional incentive would be necessary?

Again, I write a lot but I’m curious if it actually can be implemented technically or not.

Looking forward to your thoughts !

2 Likes

I don’t see the point of changing the referrer, as the initial referrer is the one that bring you to Paraswap and should be rewarded for that. Updating your link with another wallet will open abuse and a lost of consistency in my opinion.


To me it’s all depends of what is the final goal of the referral system. But I’m fine to rewarding users that bring new people to use Pararaswap at each epoch (it’s the case for other referral system like Binance, DyDx, Bybit, etc.) as it will means that referee still using Paraswap (ofc sybil attack is an issue + wash trading).

1 Like

100% agree. Thing is, at the moment I don’t think that it’s the fact to bring someone new to ParaSwap that is incentivised, neitheir in what I was thinking about, since prior ParaSwap activity is not taken account to become an eligible referee.

So the referee is not necessarily someone who has been introduced to ParaSwap, but rather someone you want to bring back regularly to build volume and share a boost.
At least that’s how it’s done today.

So there seems to be two options:

Option 1
We want the referrer to be really rewarded for bringing someone new to ParaSwap (or who has used it very little so far), at this point I wouldn’t be shocked if the referee and referrer were linked ad vitam eternam.
Example: Wallet A has done 0 or less X swaps on ParaSwap, it is considered new. He is able to use a referral link to reward his referrer.
Negative point: it seems restrictive.
Positive point: it’s the very principle of referral imho.

Option 2
We want to reward the referrer who brings volume on ParaSwap (whether the referee is new or not).
At this point you rather reward the regularity and the fact that the referrer will come back on ParaSwap. And then the permanent referral link makes less sense in my opinion. Main point of referral is to bring volume.
You can however highlight the fact of bringing someone for the first time rather than encouraging this person to come back (I would say that it is the performance of ParaSwap that makes him come back, and the social escrow if he becomes a staker of course).
Example: Wallet A is considered new and is onboarded by Referrer A, Referrer A gets a double boost X (linked to the volume of the referrer) on the first epoch because he brought back a new user, then this boost goes to simple X on the following epochs.
If wallet A has the possibility to change referrer to a new referrer B, referrer B which seems to encourage to use ParaSwap again would have a boost X/2.

And whether we go with a paraboost reward based on the volume of the referee for the referrer or any other solution, I think it would be nice to make the distinction.

At least it’s how I feel about permanent link between referrer and referee.

1 Like

Thanks @stikers for putting together a comprehensive overview of the current referral system, and its flaws. Some things I wasn’t aware of myself (like the need to use the referral link every time, which is just bad design frankly). In the interest of moving this discussion forward expeditiously, here are where I think a first proposal to address the obvious flaws in the form of the referral systemare already at a point of major consensus based on the discussions here:

  1. Linking referrals by address rather than weblink
  2. Blocking several sponsors for one address

I agree with @Xutyr with the stakeDAO example that linking the wallet forever is the way to go, especially if we’re targeting influencers. I suggested something similar in the Paraswap feature requests page, referencing GMX. This is what I would categorize as what is currently best practice in DeFi that we should emulate, and we should move forward on this change immediately.

Side note: @stenwulf marked my feature request as “in progress” so it sounds like the change towards referral links and wallets is already being worked on by the core team. It would be good to know what has already been worked on or close to delivery.

This is still a gray area in the DAO - I don’t think any minor UI/UX improvements necessarily need to be voted on, but since this touches rewards from PSP 2.0, I understand why this needed to be brought to discussion.

Where I think we are not at a point of consensus and where a vote should be delayed is in changing how the referral boosts/rewards work. There are some interesting ideas here, including a dynamic boost system as proposed by @stikers or a pro mode version sketched out by @0xYtocin. I think this should be workshopped more and should not bog down the improvement of the current referral system above, as I think this has attracted the most comment and discussion.

@Albist has also introduced an interesting distinction - do we want to attract new users, or more volume to Paraswap? The distinction here affects how we think about revamping the referral system’s rewards, and deserves deeper and longer discussion. Again I will refer to GMX here because a) they have found fantastic, lasting product-market fit that is reflected in their revenue and user statistics, as well as their price and market capitalization; and b) they have a simple, battle-tested referral system which has run more than a year now that has been immensely helplful in creating the flywheel of greater volume, more traders, attracting more volume and traders, eventually reflecting in earnings and price. Here is a quick outline of their referral system, but if you want to read more you can find expanded details in their Gitbook:

  1. They reward both the referred wallet (trading fee discount) and the referring wallet (trading fee rebate) starting at 5%. Rebates are paid in native gas token (ETH/AVAX), uncapped and in perpetuity
  2. They have a tiered system which gives a 10/15% discount vs. 10/15% rebate + 5% bonus fee in esGMX to the referral code, rewarding consistent weekly trading volume + minimum number of users
  3. The base rewards (discounts and rebates) are uncapped, but the 5% bonus rebate in esGMX is capped and paid in escrowed GMX (which need to be vested over 1 year as GMX)

There are a few basic principles which I want to expand on here:

  1. Basic referral rewards are available to everyone, regardless of volume, number of users referred, etc. 2. Upgraded referral rewards focus on multiple metrics that matter to the protocol (volume AND users).
  2. Rewards are uncapped (except for esGMX rebate), and you can refer as many wallets as you want.

What is the rationale behind this? I argue that this generous system of GMX activated a flywheel of active users who evangelized and referred others, bringing more volume and trades, making the GMX token more attractive to hold even while paying out attractive discounts and rebates to traders. As a result they have emerged from a cluttered landscape of on-chain markets to be a dominant player in terms of mindshare and financials.

In my opinion, Paraswap is nowhere near the point where should consider being stingy with referral incentives. We are, by any measure, a small player in the space vs. larger competitors like 1inch and 0x (see here, here and here for Dune dashboards showing our market share). Referral rewards, coupled with the attractive tokenomics of PSP 2.0, is a key weapon in moving volume to our aggregator, to the benefit of all PSP stakers in the long-term, as long as we incentivize the right thing. In this case, learning from the example of the PSP gas refund, I would propose a tiered, uncapped referral rewards system designed around trading activity that creates fee revenue for the DAO. As long as the liquidity is not toxic but creates revenues, we can build a business case about what share of that revenue can be redirected to the referral and the referee, similar to GMX.

1 Like

This is one of the points on which I strongly disagree.

It seems that in the crypto world, it would take huge incentives to give an interest to go or not to go on a protocol.

This is a tried and tested system, which has proven its worth in the BR, when clearing was everywhere.
However, and we saw this with the grumbling about the change to GRP. If you put carrots everywhere, people come for the carrots and not for the protocol.

Paraswap will come out on top the day they are first on price. That’s its market, that’s its goal, period. By giving incentives from everywhere, we will continue to attract people who are just interested.
After that, it will be complex to reduce the incentives even if it makes sense and we will gain, again, unnecessary fud.

In my opinion, we need to reward fairly and sustainably. Concepts that work today and will be viable tomorrow. Of course, there will be adjustments but nothing that will make us lose 12% of the liquidity in one announcement.

The heart of the social escrow is the paraboost. The heart of social escrow is revenue sharing.
Knowing this, the axes are clear to me.
Work on a coherent boost for the sponsor and the sponsored child.

No need for a free PSP, no need for a bonus GRP. A volume table that favors the biggest sponsors and a decent bonus for the referrals.

Paraswap has to be part of a logic that fits this V2. A logic that says:
Come here, you get the best prices, and you earn what’s fair… What’s fair is 125% apr at the last Epoch!!!

I’d rather allocate PSP for “marketing” and ephemeral quests, than to launch incentives that will be neither sustainable nor useful in the long run and will earn us fud if we decide to stop them.

2 Likes

I’m giving this discussion a boost.

It is clear that everyone agrees on making the referral system simpler as a first step.

The following ideas seem to have found consensus:

  • The address of the referee is linked to the address of the referrer for ever
  • The referee cannot have more than one referrer, nor can he/she change his/her referrer.

In addition, some new ideas have emerged that seem to be interesting.

  • The creation of a customisable address to simplify sharing and avoid errors when sharing the link.

The process could be as follows @Xutyr Idea :

  1. twitter authentication from the app using the twitter api. once validated, twitter normally redirects to the app where it was requested to co with an access token

  2. ethereum signature with the user’s twitter account in order to confirm that it is indeed his eth wallet

  3. verification of the signature via the ecrecover()

  4. bbd storage of the twitter and linked to the wallet (also verification that the twitter is not already assigned to a wallet)

  • Keeping the slippage rewards as they are defined today but in SEPSP1.
    Referral is a complex job that deserves to be rewarded in the best possible way.
    The idea of sharing the positive slippage makes sense but the need to claim several currencies is not attractive, nor financially interesting.
    The solution found by @Xutyr would be, like the Gas Refund, to allow referrer and referee to claim their slippage rewards in SEPSP1.
    By doing so, we avoid claims that make the slippage sharing measure useless for small amounts and we favour the staking of the users concerned.

IMPORTANT : Concerning paraboost, the data in our possession today does not allow us to make an effective projection for the realisation of paraboost levels.
Reflections are continuing and should in our opinion be the subject of an independent proposal.

1 Like

I agree with these points.

I think it’s way to complicated, the process to onboard new users should be effortless.

Can you please clarify your idea :

  1. 100% of the positive slippage would go to the sePSP (profit sharing) earn and $PSP from the DAO treasury would be spent to reward referrer/referee?

OR

  1. All tokens from positive slippage are cumulated in a dedicated wallet, swapped for $PSP (buy back) at the end of the epoch, locked in sePSP1 and the distributed/claimable by the referrers/referees?

If option 1) then that means additional selling pressure.

Why doing that in another proposal, I think it’s the most important thing to work on.
The problem is that you only need 1 trade, no matter the volume (even $1) to unlock referee and referrer boosts, easy to game and not incentivized to generate as much volume as possible.

I think that we should establish the same kind of rules as for trading boosts and think if it should be :

  1. from 5% to 30% like it is for stablecoin to stablecoin :
    image

OR

  1. from 10% to 60% like it is for non-stablecoin trading :
    image

For the moment, 50% of the positive slippage goes to the referer and the referee (25% each). 25% instant for the referee as positive slippage and 25% for the referer at the end of epoch

The idea was to keep this but, at the end of the epoch, all the slippage that should go to the referer (usdc, usdt etc depending on the positive slippage of their exchanges) is converted to PSP (buyback) then sepsp1 which can be claim by the referer.
We keep the current system while creating a buyback on the PSP, by creating only one claim instead of a multitude.

The problem is that by taking the existing volume tables, we are ignoring the number of referees.
The person who sponsors 100 people will quickly reach important levels which is not coherent in my opinion and will not meet our need to mitigate abuse.

1 Like

It’s not the process to onboard new users that has been describe by stikers but the process to create the referral link. The idea is simply to share for you referral link something like: https://app.parraswap.io/r/Enerow instead of https://app.paraswap.io/?ref=MHg5OWUxODIEGHB55zaaef58mMzMTlhMTZlYkQ2YzIyNGY0
and when the user doing his first swap with a referral link he will be link to the referrer for ever.

2 Likes

Option 2) then.

About sePSP1, I think we should put ourselves in the shoes of referral professionals/influencers/etc…, we need to propose what the industry or even better to be attractive.

Do we think that sePSP1 rewards are attractive? I think first we should do some research/thinking about what would bring professionals/influencers/etc… in the ParaSwap Referral Program. I don’t have the answer but I doubt it’s sePSP1.

Why not rewarding them in stablecoins or eth instead?

I don’t know if we care about the number of referees, what counts is the volume generated by these referees. At the end it’s the volume that generates revenue.

Ok, thx for the clarification. But anyway, why do you wan’t to use twitter to authenticate while we’re in a web3 environment? We can still propose to create user readable referral links without having to integrate twitter API don’t you think?

So what we care it’s positive slippage volume and not volume, in that case I’m agree to have rules equal to trading volume. But having the same rules just with volume will create so many fake volume without value from whales. Wallets that exploited GRP will come back with a strategy close to:

  • Create a main wallet will a good sePSP position and do 100k volumes on it to get the easy level for whales (15% mini)
  • Create a new wallet using referral link and make the volumes

So a single user (whale) can generate 0 new protocols revenues and increase its yield (and having gas refunds at the same time)

As sePSP token it’s a ERC20 you can repeat the process at each epoch with fresh wallet.


It’s mainly to avoid fake account promoting link from famous people. Imagine a fake twitter account with a custom link of https://app.parraswap.io/r/ElonMusk


The idea of a referral system should be to bring new users not new wallets own by current users that just want to exploit the program for more yield.

1 Like

The idea is to have a PSP buy back. If we consider that a person is ready to promote Paraswap it is because they are :

  • Agree with the idea that paraswap is a successful platform
  • Agree with the social escrow system and therefore personally involved in staking
  • Satisfied with growing their bag passively via referral rewards

Can you imagine? Hey, I refer you on Paraswap but I don’t stake there because it’s shitty?
I can’t believe in this speech but maybe my reasoning is too simplistic
In the worst case, we are talking about an additional claim and 28 days of unlocking to take advantage of its PSP to sell them?
Is this a problem? as we have simplified the way of claiming

The gas refund is a tool where we offer PSP. We also need to think of attractive tools with passive purchase of PSP.
The referrer buys PSP without doing anything, no impact for him, only gain and a positive impact for Paraswap.

In my opinion, it’s all in the way you communicate the solution.

Either I don’t understand what you’re trying to say or you don’t understand what I am saying.

How creating 10 siblings wallets doing $10K volume would give you more advantage than by only 1 wallet doing $100K volume?
At the end the total referral volume for the epoch will be $100K anyway so same referral boost level.

Yes I can imagine, that’s how it works, referrers are chasing for the best offers/earnings not especially for the best product and usually are not here to accumulate tokens but to get an earning out of their audience.

1 Like

Yeah you’re right, mb. Updating my previous post

So you think that we should give weth. We’ll it’s make sense too. I’d like the idea of buyback but if it’s not interesting for the referer so…