A Roadmap for a Sustainable PoolTogether

[Brendan]

A Roadmap for a Sustainable PoolTogether

In a previous post I described what a sustainable, decentralized PoolTogether could look like and the feedback was overwhelmingly positive. Here is the associated Snapshot Vote

In this post I’m going to outline a roadmap that describes the steps we need to take to achieve that vision. This post is presented as a strategy for the protocol. This is not a strategy for any one team. We’ll start by looking at our finances and how we can manage our burn rate, then we’ll look at what we need to build.

Managing Our Burn Rate

The PoolTogether protocol treasury has ~$1.93m in assets remaining. Our goal for these assets is to build a successful prize savings protocol that grows on its own accord and benefits the POOL token holders.

Given our finite resources, we need to carefully manage our spending so that we are able to set the protocol up for growth. It will be much easier for us to strategize our spending by thinking in terms of a framework.

A Spending Framework

We can simplify the construction of a protocol into three key phases:

1. Launch. The first complete deployment is launched on a chain.
2. Replication. Once the deployment has proven successful and has matured, it can be forked to other chains so that it can tie into their ecosystems.
3. Maintenance. After a deployment matures, the effort around it is focused on business development and partnerships. The goal is increased integrations, incentives, and promotion. The protocol serves as a foundation.

These three phases require different skillsets. It’s analogous to the construction of a building; the architects are part of the design phase. The construction team is the building phase, and maintenance workers manage upkeep after the building is constructed. Each phase requires a vastly different skillset and budget.

In that same way, we should be thinking about the construction of the protocol.

Launch requires the biggest lift; it needs specialized expertise to make it happen. This will form the bulk of our spending.

Replication is far less expensive, as there is far less R&D that requires developers or auditing. However, there may be minor modifications or integrations required to function in a new ecosystem.

Maintenance is the least expensive. Someone just needs to keep the lights on and the doors open.

Our Biggest Expenses

By far, PoolTogether’s biggest expense is people. We need people to help us to achieve our vision; whether they are developers, designers, business strategists, community builders, or otherwise. People are what drive the evolution and adoption of a protocol.

To efficiently manage our biggest expense, we’ll need to match the personnel to our phase of building. However, personnel aren’t some resource that can easily be increased or decreased.

• Quality expertise is hard to find and retain.
• Our protocol requires knowledge workers; getting them up to speed with our technology and culture takes a tremendous amount of time and energy.
• Once people leave, they don’t come back.

This means we need to be very careful with the people that we have, and selective with who we bring on.

Conclusion

The bulk of our costs at the moment are builders; and for the first two phases we need to retain them. In this way, we need to front-load our spending. We should spend liberally early on to achieve launch and replication.

The V5 core took about eight months, from start to finish. It was a complex project, and required heavy auditing and testing to ensure that it worked successfully.

I think we can safely estimate that we can achieve full re-launch of the core with new pieces, and replication to more chains in less than six months. After this time we can reduce our personnel according to our needs.

The Three Phases

Phase 1: Launch

We finalize a deployment of PoolTogether that forms the foundation for our future vision of PoolTogether and addresses any shortfalls or technical issues we’ve identified since our initial launch on Optimism.

Tokenomics

We need to establish the narrative for the POOL token. We should renounce ownership of the token so that no more POOL can ever be minted. This reinforces the finite nature of POOL, while leaving the protocol some flexibility thanks to the POOL in the treasury.

The treasury has about 40% of the supply of POOL tokens (~4m), which can be burned or disbursed later.

Incentives

The treasury holds OP tokens, which naturally lead us to deploy on Optimism and incentivize the vaults with OP. This will help bootstrap our first deployment.

Assets

We forge ahead with the USDC and wETH assets.

Additions to the Protocol

POOL Staking

With a finite supply of POOL tokens, the only indefinitely sustainable and scalable source of staking incentives would come from the Prize Pools themselves. A portion of each Prize Pool’s yield will be set aside for stakers on that prize pool.

Prize Pools will be autonomous, so it’s important to have this as part of our launch.

Prize Compounding

The ideal user experience would be to win more of the tokens they already hold. This is achievable, but complex. The POOL prizes would need to be efficiently and trustlessly converted back into the asset of whatever vault won.

Since the protocol is permissionlessly extensible, vaults that behave this way can be added later. We need to evaluate whether automating this is worth the effort for launch. To test the viability and desirability of prize compounding with low effort, we can introduce interface improvements that enable users to compound their winnings manually by directing them to a POOL to PUSDC/PWETH swap.

Needed Improvements

Several existing shortcomings need to be addressed:

New Vault

The old vault has a collateralization issue that is problematic for users and prevents us from growing.

We need a new vault that is compatible with Aave and allows us to grow indefinitely. This vault needs to be audited and a new factory deployed.

More Efficient RNG

We need to minimize the RNG costs so that the reserve and stakers earn as much POOL as possible, otherwise staking is ineffective until higher TVLs.

Prize Pool Tweaks

The prize pool allocates too much liquidity to the daily prizes when tiers are reduced. This needs to be adjusted.

Phase 2: Replication

Once we have the protocol running successfully on one chain, we can leverage the traction to build out the protocol on other chains, with new yield sources, or new assets.

We must be agile, and seek to deploy where:

1. There is longevity. When we deploy we should be thinking long-term.
2. There are incentives. If we can arrange incentives for the users, it will drive traction.
3. Effort is minimal. We must shoot for easy targets first, otherwise we risk spending more effort than is worth it.

Ideally we’ll be able to simply re-deploy the protocol, but there may be instances where we need to integrate with new protocols for a more “native” implementation. Custom yield sources, RNGs, or otherwise may prove to be different.

Phase 3: Maintenance

Once the protocol is live and running, we can reduce our technical expenditures and focus on community and partnerships.

• Wallet integrations
• New vaults
• New yield sources
• etc.

It will be important for us to continue to have a place to bring users and contributors together; people need a place they can come and ask questions and explore partnerships. We will need to maintain a funnel for these people.

We will likely still need a technical arm along with a budget, but it will be for evaluating integrations, funding audits, and require minimal custom development (ideally!).

Beyond Phase 3

If this plan works out, the POOL token value will increase and there will be an opportunity for POOL token holders to capture value. The token can be leveraged in numerous ways.

Treasury Burn Rate

The following chart shows the treasury levels against a quarterly burn rate. This excludes POOL and assumes no more funds are raised by the treasury. The total treasury is the y-axis, and the quarterly burn rate is the number within each bar.

The intent of this graph is to show how we can manage our burn rate to front-load costs and retain people while still getting good longevity out of the treasury. These numbers are meant to be illustrative and not taken literally.

If we front-load development now and progressively lower our burn rate, we can easily support the protocol for another two years.

Summary

I believe we need to employ the above strategy to set PoolTogether up for success. We will front-load the development and construction of the protocol so that we have a foundation on which to stand. Once we have a strong foundation, we can reduce our technical efforts and focus on helping others build on the foundation.

I’ve created a Snapshot vote so that POOL token holders can signal whether they agree with this strategy.

Snapshot Vote

If anyone has any feedback, please comment below.

[underthesea]

Alright, Im here for the gov forum points

Sorry but I do not like the snapshot. It’s confusing what we are voting on exactly. Treasury spending happens with POOL votes. I think we should discuss more.

It sounds like the sustainability is in the POOL staking, a direct cut on yield to stakers. Lots to discuss there, not my style, but Id be interested in talking more on the particulars. What percentage? I assume this is on the prize pool because otherwise it can be competing against?

We forge ahead with the USDC and wETH assets.

I strongly believe we are missing our potential here. We are a lego on yield sources, having only one is very limiting. We need to make this easier and more inviting. We need to add more yield sources ourselves. We cannot ignore LSTs now. Yield sources are a key to growth. PoolTogether core functionality to me is yield to prize. More yield more prize. POOL has become a lot of the focus. POOL does not make prizes.

The treasury holds OP tokens, which naturally lead us to deploy on Optimism and incentivize the vaults with OP. This will help bootstrap our first deployment.

The PoolTogether protocol treasury has ~$1.93m in assets remaining.

Assuming the $1.93m includes the OP but we seem to have a double spend here when looking at the treasury spending plan and using the OP for incentives.

Prize compounding - I don’t think it’s worth the lift to put a swap widget in the app as a means of user testing. There’s no question in my mind that people will want to win their tickets instead of POOL. To minimize costs of this swap-back I believe we need to raise the value of the minimum prize relative to gas. This way the fixed costs of swaps being added to the claim cost are a lower percentage of prize delivery. Also people don’t like penny prizes.

I like all three improvements outlined - new vault, more efficient RNG, and prize pool tweaks

I like the focus on one chain (Optimism) until its success is proven

I agree with the shooting for incentives on other chains, looking for longevity, and minimal effort. I’ve participated in deploying V4 on Optimism and the code was very well setup for near play button deployment.

Overall I think the spending idea of more now less later is really not negotiable given the state of things. I do not think $500k this quarter is starting on the right foot. Would like to see it reduced.

editing to add that in the spend and sustainability planning liquidity is a big part of the equation. or maybe not, but we should make that explicit if not.

[Taliskye]

[Taliskye]

Y’all I think we’re in the worst PT timeline. We as a community came together and voiced concerns on over spending, lack of direction on deliverable, and lack of accountability for what’s been accomplished (and lack there of). And the response is a spend budget that would only cover costs of Generation Software with a nice ramp so they can prep dissolution or ramping into another project. Has Pooltime been consulted in any of this?

This might go long, but I’m going to be short and sweet. I can’t believe this.

[edit removed non constructive feedback]

This is a spend plan to move treasury to Generation Software. This is not a vision or a roadmap.

For those not on the council call here’s some take aways:

• We need to pay for the vault fix.
• That fix should be “a pretty simple fix”
• That fix needs to be audited.
• That audit cost ~50k… but that’s because audits cover a certain amount of work, so Generation Software needs to package other contracts into the audit to get the most bang for the buck.
• The community is free to enlist some other dev-shop to make and audit the vaults.
• Basically everything the Pooltime team has lined up for BD and marketing is dependent on working vaults

Take the tone of those bullets to be less harsh, I’m not in a position to word them in a totally friendly, non-curt way. I’m floored.

[TheRealTuna]

I voted yes to the first snapshot, it didn’t really outline much but I do want a sustainable Pooltogether so I figured why not. I was concerned about the plans over a year ago and have voiced them along the way but still let things pass as I wanted to give it a chance and I can’t control what gets built.

This plan feels like more of an outline of why we’re gonna have no money in a year than a plan on how we are gonna build the perfect hyperstructure and scale it up to the TVL needed to explode before money runs out. Things like liquidity, depositor incentives, and POOL staking need to be in action at once. The way things are being done it feels like you will work out each thing one at a time and we will never have a system firing on all cylinders.

The purpose of Pooltogether was to forgo a negligible amount of interest for a chance to win a life changing prize without losing your principle. This was a great concept that earned the respect of Vitalik Buterin himself. This system requires a gambler to save more money to improve their odds. Instead, the system we are spending millions on takes negligible interest and spits back out negligible interest when it’s working. I understand the vision that when we reach a certain TVL it will put immense pressure on the token to go up and the token and prizes would grow huge, but to me it feels more likely that we will eventually get working vaults when we are near out of funds and have nothing left to incentivize TVL growth to the needed level. You have gone the route of building something so complex that we will be out of money before it’s complete.

The above table literally shows us handing over the treasury to the team and then trying to sustain the protocol on $50K per quarter for 18 months.

I think you are missing the point.

[Brendan]

There are misconceptions here that I need to nip in the bud!

@Taliskye

This is a spend plan to move treasury to Generation Software. This is not a vision or a roadmap.

This post is written from the perspective of the entire protocol. I’m putting my protocol leader hat on. When I throw out the ballpark $500k burn for a quarter it’s for all expenses to the treasury. This is not a PTBR for GS.

On that note, the numbers I provided are ballpark numbers; I made them up. The intent was to communicate the strategy of front-loading the costs for our burn rate. These numbers shouldn’t be taken literally.

@TheRealTuna

handing over the treasury to the team and then trying to sustain the protocol on $50K per quarter for 18 months

Each bar in the graph shows the dollars spent for that quarter; that’s why it’s $500k for the first two quarters, then $250k, then subsequently lower and lower. They’re meant to illustrate the front-loading of costs and what the complete burn rate would look like given an approximate spend. Take another look at the graph.

[TheRealTuna]

Yes, it does appear i misread the graph. It is still a tough pill to swallow that we have to trust this plan to work out and if it doesn’t we are left with nothing. Pooltogether is such a great concept to turn gambling into saving. A bit ironic that we as a community are stuck in such a gamble.

[AndyKaufman]

@Brendan

In a previous post I described what a sustainable, decentralized PoolTogether could look like and the feedback was overwhelmingly positive. Here is the associated Snapshot Vote

Being aligned with a feel good potential scenario does not mean being aligned with present actionable items…

[mr_DvD]

@Brendan

The PoolTogether protocol treasury has ~$1.93m in assets remaining. Our goal for these assets is to build a successful prize savings protocol that grows on its own accord and benefits the POOL token holders.

Thank you for removing $POOL tokens from the equation, at the moment at least, since we are really low on liquidity their real value is much less that what a naive reader may read.

I do agree with the previous concerns but we have chosen a road and we cannot U turn back completely: at this point we only can do damage control and get to the end of it.

The end of it is not that bad at all in my opinion, we did a lot of great things and decision but also few bad decisions that are still relevant today.

I’m not going to repeat what @underthesea, @Taliskye and @TheRealTuna because I agree more or less on everything.

I only see fees much more needed than @underthesea, but I think we all are aiming for the same destination, just with minor differences on the road to take.

Since I’m already wearing the unpleasant guy hat let me go on with that role, but remember that in business if we are only kisses and flowers and never point out the problems we are not lasting long.

The problem is that we became irrelevant and boring.

Who is going to be excited for a $0.20 prize?

Do you really see people saying: “I’m going to deposit my $50,000 on PT, because I will win $0.20 often! And may more rarely $100”?

We need to attract the big fishes. If I had $50,000 I would put them on AAVE and farm 8% APR on STABLES. That is $10 daily.

So if I had to make few moves minimum cost max impact they would be:

1. Rise that average prize size, but rise it A LOT;

2. Instead of subsidizing TVL with OP, use OP to subisidize the grand prize we are waiting and dreaming a big grand prize when we could make it happen today with a fraction of the money we are wasting in incentives we already know are not working ( cfr: wasting $1.0M (ONE MILLION DOLLARS) on irrelevant prizes on V4 really taught us nothing?);

3. Become less boring for the big fishes: we need whales to deposit and eventually buy pool. We are all crypto natives we know what a whale wants, what are the whales language.
   Our main community hub is Discord and it’s the less attracting place for whales. Community job is to attract capitals on TVL and on POOL.
   Whales are on Telegram. I hate telegram too, but it moves a lot of money.

We need people talking organically about the prize they can get here, now. We need people talking organically about how cheap is POOL now.
The key word is organically. Paying people to make paid content will not work: we would pay for great content that no one will consume and that will not create an impact.
Diverge those money to the grand prize instead.

All in all I think we are on a verge: we can still make it but this is the last chance we have to change route.
We either change our sentiment about money, revenues, and general openness to wales and human nature, or we will fail out of irrelevance and broke in 2024.

[BRONDER]

I’m going to be “short” here.

To me it kind of feels like this is a very similar post than the TBR GS posted. Basically trying to ratify that the items proposed in that TBR are the direction to follow (there are some differences too, for example NFTs are not mentioned in this post).

As I said before, I don’t think those are the most important points for the protocol to develop. We are at a situation where our TVL is pretty low. Without yield, prizes are low too, and less people are attracted to deposit.

Do I think POOL staking is going to change this situation? NO

Do I think prize compounding is going to change this situation? NO

This does not mean that I am against those 2 previous developments, it’s just that I don’t see them as an urgent matter.

So, what is urgent then?

Obviously we need to fix the current vaults so that we can start on the right foot. While we are on it, it’s also a good time to reconsider prize structure in case we have learned something from these past months.

After that is done, we need yield sources. With the appropriate visibility and the right yield sources, we might have a chance to grow this thing and get a big prize that can attract people.

[DaBoom]

Echo Bronder’s feedback and the rest of the responses’sentiment. We were against the TBR because of the cost for product. This post is further justifying that cost. The community has spoken. Let’s fix the urgent items and go from there.

The Treasury cannot be responsible for salaries. The Treasury is responsible for the items voted in favor of in the Proposals.

If a proposal goes live to fix the vaults, I vote yes (for a reasonable price). The rejected proposal had deliverables that are now being described as “simple fixes” and “items that are much cheaper to build” yet the size of the cost was the same as the previous ones and bigger than the ptinc initial proposal to build and launch V5.

Let’s fix what we have, and then go from there.

[underthesea]

For the direct tax on yield directed to stakers, Im still surprised by this. I thought V5 was designed specifically to create value for the token that was not so obviously rent seeking. If we are just going to turn the fee switch on then the whole POOL buyback becomes a relatively less efficient fee where revenues are lost to gas.

I’ve said this before but there are paths to have an aggregated grand prize across deposit tokens without using a unifying prize token.

From my perspective it is one fee or the other. We can’t be frivolous with taxing yield. It should be a direct fee or the buyback, not both. If Im choosing one, I think we go with the buyback as it stands now. In my vision neither is necessary nor long-term competitive. Fees being a race to zero.

[Brendan]

After the community call yesterday it was clear that there was still a lot of confusion around this post. I want to include some straightforward answers to peoples questions:

Here are some key points summarized:

I estimate we can achieve phase 1 (launch) in one quarter, and phase 2 (replication) in the second quarter.

The burn rate is a projection, the numbers shouldn’t be taken literally. It’s meant to illustrate how we can front-load spending to retain talent now while still having longevity from our treasury.

Some people are still unhappy with the details of phase 1 (launch). There is still plenty of room for change in phase 1. We’ve been listening to the community, and are going to kick-off a conversation soon around some paths forward.

[links]

I was not able to stick around for the community call yesterday, so I’m sorry if I am missing some points, but I will leave my feedback anyway.

It’s extremely hard for me to judge this plan based on the words of the post and the past PTBR. It’s mainly because the words are talking about costs and not revenues. I know that it costs money to build a product people want to use, but without putting those costs into context (i.e. it’ll cost us $1M to build a product that generates $10K monthly for the treasury), it’s hard to me to tell whether or not this plan is “good”.

You keep using the word “sustainable”, but it sounds like we have different definitions of what that means. It sounds like you are saying if we build a hyperstructure, the product will live forever on the blockchain and our job is done. To me, sustainability means that the protocol can continue to thrive (not just exist) in perpetuity. For the protocol to thrive, we NEED humans working on it. We need to pay for people to evangelize the product, help new users onboard, fix bugs, make updates, etc. It doesn’t feel like this plan takes this into account.

If you are going to write a post from the perspective of the protocol, I suggest strongly you create a spreadsheet (cash flow) which shows costs and revenues. This will force you to make assumptions about things like # users, average deposit per user, costs to onboard users, etc. When all of these assumptions are written down, it makes it easier for everyone to align on whether or not these assumptions are true. Here’s an example I did for Bankless Card: Bankless Card Business Model - latest - Google Sheets

PoolTogether has a lot going for it - the passion of the community is a good thing. I hope you’ll work to get the community’s support, because we’re more likely to win together.

[Brendan]

@links

You keep using the word “sustainable”, but it sounds like we have different definitions of what that means. It sounds like you are saying if we build a hyperstructure, the product will live forever on the blockchain and our job is done.

No- we both agree here. This is what I said in the Sustainable PoolTogether post:

A protocol is nothing without its people; the stewards of the protocol.

It’s important that there is funding for development, bd, marketing, design or otherwise. Each chain, asset, or partner will have their own needs, so decision making must be decentralized in order for it to scale.

Stakers can choose to keep their rewards, or band together in groups to share spending for the efforts above. These groups will coordinate via multisigs, token-weighted voting DAOs, or any other means.

Ultimately we all want the POOL token to be valuable. That is without question something we all agree on. The sustainability post describes what that will look like in a fully decentralized future.

No matter what we do to achieve sustainability, revenues aren’t going to immediately exceed spending. The point of this post was to communicate that we need to burn cash to try to reach sustainability. That’s a fact- we’re going to spend money faster than it is made. Our spending strategy is critical; I raised this point because if the token holders squeeze the budget too early, then we’ll lose talent and we won’t have those people that are so important.

I think what you’re wondering is what do we spend our money on, and what are the projected returns for that investment. I think that’s a different question, but it’s very much related to the content of the phases above. I regret including the details of each phase because I think that has muddied the conversation.

On that note, the GS team met today and discussed the feedback from the community regarding the PTBR and everything else that has been said. We’re going to present the paths forward as we see them, including a recommended option. I’ll make sure to include projections for POOL staking based on TVL, so you have some data to work with.

[mr_DvD]

@Brendan

Ultimately we all want the POOL token to be valuable. That is without question something we all agree on

I’d be surprised if we all agree on this.
I do agree.

[Taliskye]

@Brendan

There are misconceptions here that I need to nip in the bud!

Appreciate you calling me out here. I definitely approached this from the perspective the PTBRs will not change significantly enough to adapt to the burn rate explained here. I look forward to seeing the updated PTBR. My personal head cannon has GS 60-75% of the budget laid out.

@Brendan

I estimate we can achieve phase 1 (launch) in one quarter, and phase 2 (replication) in the second quarter.

I missed the call, I look forward to listening to the recording. But I have trouble with this timeline, there’s budget requests going back to 2022Q4 for researching v5 and three quarters 2023 Q2, Q3, Q4 talking about launching v5. The track record for accuracy is a surjective relationship.

[AtomicNuclei]

@Brendan

When I throw out the ballpark $500k burn for a quarter it’s for all expenses to the treasury. This is not a PTBR for GS.

Okay, that’s a really important call-out. I don’t think that was apparent from the post and caused another little uproar, lol. I like the direction we’re going in finally, consensus building! phew. :3