INDEX.labs - an idea to consider

Should the coop set up a more experimental approach to launch simple funds?

Something like the ape.tax from yearn, or Alpha’s Innovation launchpad for methodologists

This is not an original thought, I know a few people have been discussing it at different times and the lack of a launch to date indicates that the consensus has been to delay. But the idea keeps coming back to various individuals.

With the acknowledgement of the current constraints on Engineering resource I think it’s worth opening the discussion to the smartest people I know: INDEXcoop

Context:

As the Product Working Group Lead I’m very aware that we have a number of excellent proposals for products and that are stuck in a limbo between the initial proposal and the launch to market.

These include:

Product Status
TTI Failed DG2 quorum
dVIX (volatility) Pre DG1 discussions
SYI (Stable yield) Pre DG1 discussions
iRobot / PMI Pre DG1, back test
Category Winners Draft proposal
Uphold Defi Draft proposal
DAO Lego Sentiment check post
Sentiment (SODA) Draft proposal
DATA Draft Proposal

Whenever I see these posts, I’m impressed by the thought and energy that has been put into them and I have a number of questions:

  1. How complex is it to build and maintain?
  2. How big is the market / do we have product market fit?
  3. What is my gut feel about the product?

#1 is easy to estimate, some products are are very complex (FLI), some are no standard, generally to capture income (SMI), others are simple collections of governance tokens with periodic rebalances via swaps (MVI and DPI).

#3 is purely personal on the product proposal.

#2 is in many ways the most difficult question, How much will people hold / trade?

We have developed tools to assess products as part of the process from idea to launch, but Product size / market fit is still (at least for me) the great unknown. We are learning, and Punia presented a great group of insights gained into our products here:


So, what is the goal of the coop?

Grag’s excellent post summarised the coops Q2 goals as:

With those learnings downloaded, let’s look again to our Q2-2021 pursuit of 4 aggressive growth goals:

  1. $500M AUM
  2. 400,000+ DPI supply units
  3. $9M FLI suite 7-day moving average volume for 2 consecutive weeks
  4. 4 new products launched with $200M+ combined TVL. They are category leaders in terms of the most relevant measure (i.e. uTVL, volume, liquidity)

With the engineering challenges we face, full coop product launches are going to be slower than we might like.


A potential INDEX.Labs goal:

To improve the coops process to help identify sector winning products for full scale launch by running smaller on-chain funds.


So, how could it work?

As a start, I would suggest the following as a Minimal Viable Product:

  1. Methodologists agree that they want their proposal to be considered for the .Labs.
  2. Coop community decides on 1 (or 2?) proposal to launch within the .Labs.
  3. We communicate the planned launch date and run time (6 months?)
  4. We open an address to allow users to contribute ETH.
  5. After the agreed period we do the initial swaps and issue the tokens via the multisig to convert the pooled ETH into the fund token
  6. Then we manually distribute the fund token to depositors.
  7. At the end period, we either unwind the product (convert back to ETH?), or we migrate it to a full product.

Potential parameters:

  • At least 50 ETH deposited before we launch.
  • 400 ETH supply cap
  • 6 month runtime
  • White listed addresses / INDEX holders / threshold only

Coop resources required:

  • Minimal possible Eng support to set up multisigs, create the Set, training on rebalance etc.
  • Community members to hold multisig to participate in launch, distribution, and rebalance.
  • ETH for gas
  • [optional] INDEX as a stipend for the methodologist and multi sigs.
  • [optional] ETH to co-invest in the fund to build AUM
  • [optional] ETH to co-invest in the fund and then create and seed liquidity to Uniswap to allow trading.

Questions for the community:

  1. What would be the goal of such a programme?
  2. What implementations should we consider?
  3. What are the pro’s and cons of doing this?

And eventually, do we want to advance to a formal proposal to do this?

10 Likes

I have to admit, I don’t yet have a strong opinion on this idea.

Benefits:

  1. We will learn something about whether people want to invest in these products.
  2. We will learn about the performance of the products in the real world.
  3. Our methodologists will see a clearer route to the full scale market.
  4. We will spend time thinking about the overall “idea to product” process and likely improve it.
  5. We have something to push via our different marketing channels to demonstrate that we are innovating and searching for product market fit.

Downsides

  1. I think the biggest is it becomes a trusted set-up: deposit to a multisig, wait a week (?), get tokens, and rely on methodologist not to screw up the rebalance. The coop / Set have a whole bunch of safe guards and a plan to decentralise and I suspect any INDEX.labs is going to be a centralised solution.
  2. Do the multi sigs become financial service providers / personally responsible? Would I have the FCA / SEC asking me awkward questions at some point the next time I fly anywhere. I’m not sure I want to run that risk. There is a reason Set / DeFi pulse set up the coop… If our engineers are asked why they did trade x in a rebalance they can say - I’m just following the DAO’s instructions, I didn’t advise anyone.
  3. Loss of focus for key contributors - MVI takes a lot of attention from Paul and AG (Two of our most active contributors), would their time be better spent in different areas of the coop? - I don’t think so, but what happens when all the golden owls start running INDEX.lab funds as a 3rd job?
  4. We could get even more proposals for new products…(how many people are running set contracts as a communal investment pool?) We could manage that by having a choke on the pipeline - 1 INDEX.lab product per month - selected by vote / committed ETH / deposited ETH etc.
  5. More contracts, more multisigs, = more ways we can damage our reputation.
  6. Does a small test product, without incentives, tell us anything about how a full scale product would grow?

I suppose the real question is:

Is this the best use of our resources at this time?

@DefiJesus @gregdocter @DarkForestCapital @verto0912 @catjam @mel.eth @Kiba @LemonadeAlpha @BigSky7

4 Likes

Could this be an opportunity for a collab with Sushi on Miso?

Personally I really like this idea, but in the context of the DAO we generally like to see ourselves as more professional and buttoned up, whereas ape.tax really leans into the distilled ethos over at Yearn. Is there a way we can keep this more Index Coop and still be ‘experimental?’

We don’t have to intake whacky degen stuff, necessarily.

The idea of “Index Labs” has popped into my brain too at various times!

However, given our engineering/product constraints, I am having quite a hard time thinking that right now is a good time to use our limited resources to launch/support a brand new program.

This ^ is hit on in OA’s downsides reply – and I think those downsides are big at the moment.

imo, this might be more of a “later” sort of thing, as the benefits make loads of sense.

4 Likes

This will be a huge benefit for our culture of experimentation. If we’re too afraid to fail we are doomed to fail. This separate environment allows us to move fast and not break things.

Even if today is not the right time to start this, I definitely want to see this in the near future.

6 Likes

I like the fact that allowing customers to try our products can let us know if the Unique Selling Benefits (USB) we think will attract customers is something they feel they actually get from the product and if they are willing to pay for it. I believe it would let us evaluate multiple aspects of business strategy before coop commits to a larger budget.

Testing might reveal that you can drop a particular feature and save money since consumers don’t need or want it. We could also get unsolicited and solicited user feedback to help coop tweak what we are selling or how we are selling it, to maximize profits and minimize risks.

2 Likes

Hi Greg,

I will admit, the technical requirements are a challenge.

However, I think that required skills:

  • Set up multisig
  • Swap ETH for components in correct proportions
  • Create token set
  • Issue all tokens
  • Rebalance trades within the Set.

Are relatively simple, (you don’t need to write code, but may need to interact with the contract via etherscan?) and could be done by a wide number of the coop community after some training sessions.

I think the main challenge is that we would need to run a token set via a multi sig, I’ve never done it, but I think some of the social set traders will be doing it.


@dylan @ncitron @richard @puniaviision : Am I correct in thinking that the above can all be done by community members who are not full blow solidity devs?

If so, how many Gold / Silver / Bronze owls do you think could step up to the challenge with training?


My half created thought would be that the multi sigs would include:

  1. Methodologist
  2. Golden / Silver owl who is familiar with trading.
  3. Community owl who has web2 / web3 experience (not a full code Dev)would real
  4. Member of Set Eng team / experienced Coop Eng

My hope is that #4 would never be used, but would be useful to support any tricky operations.

4 Likes

Largely agree with @gregdocter here – love this idea, though now doesn’t seem like the right time because of eng/product constraints. Once we are able to focus outside of strengthening our core product set, we should pick this back up.

Would love if a future version of this minimized the eng support even further… perhaps we have some well-written documentation on Set creation, etc instead of training sessions, and have a support channel that includes all Labs members + Set devs (idea being that Labs members could also help each other!)

Might also be cool to pair a Labs team with some informal advisors from the Coop, if they’d like… we have so many great DeFi minds in this community that could act as an “advisory board.”

So many cool possibilities here, and thanks for getting the conversation going!

3 Likes

Hi everyone,

The Index.labs concept makes sense assuming adequate engineering resources. Given today’s resources here’s a semi-baked concept: an “engineering light” version of Index.labs, or perhaps it is the first of multiple stages for new products that are incubated through Labs, is an emulator that allows approved proposals to be “launched” in emulation or “prototype” mode (no actual assets behind the prototypes) that allows the Coop to observe performance but also test assumptions - e.g. does the prototype index perform as expected during periods of high volatility - that would be useful in defining the requirements before it advances to the next phase of a Labs project (e.g. “beta”) or onto a fully commercially available index.

The emulator also has the benefit of being able to absorb a higher number of proposals for purposes of engaging potential methodologists (I count myself among these) and add to community or even customer engagement, perhaps even attracting new engineering talent to the Coop. The list of 10 proposals in OA’s post could be, for example, “Class 1 - Q3 2021” and additional classes could be launched periodically. Visibility into performance might be for Bronze Owls and above (?). TBD how we determine which proposals are accepted into the emulator.

The emulator will need flexibility to accommodate a potentially wide range of index features and will most likely require a bespoke application to configure, launch, and maintain each prototype index. Happy to work with anyone interesting in scoping out requirements/specs.

1 Like

I think the products will have real assets behind them but with a huge “experimental” warning. The emulated test you describe is done during development, having real value on the product also incentivises exploits for attacker - thus increasing the security of our products.

2 Likes

This can already be feasibly done by creating sets in the builder but without minting units iiuc

5 Likes

@overanalyser If I understand correctly, your intention with INDEX.labs is to make it easier to launch simple funds (i.e. simple sector/AUM indices like DPI, MVI, or Data Economy Index (DATA)).

I left a detailed comment on @puniaviision Engineering and Product Constraints challenging the assumption that Set Engineering resources are actually a requirement for launching a simple index - I do not not believe they are.

I manage my a chunk of my personal portfolio with a Gnosis Safe multisig as a Social Trader on the TokenSets platform (see the BEST Portfolio).

Based on my current understanding, my view is that we can launch simple index products by having @dylan and the Engineering Working Group transfer rebalancing responsibilities to this group that you have outlined:

@dylan @ncitron @richard @puniaviision Please sign me up for whatever group is working on unblocking launching simple index products! I am happy to document processes, do manual rebalances, etc.

3 Likes

Not only, but also. This proposal also helps us test very complex products in the real world.

Glad to see someone trying to champion this. Simple products may not offer much competitive advantage, but they feature very low maintenance and upfront costs. If one of these simple products hits product market fit it might bring significant AUM.

4 Likes

This is an idea that has come up several times - and I like it a lot.

Over the past month a theme that has come up over and over again when talking with investors is our need to increase the efficiency of our product releases especially for low engineering lift Indexes. We have the opportunity to quickly capture the market for a number of indexes (note these have all been specific asks).

  1. DEXes
  2. Lending
  3. Derivatives
  4. Infrastructure Index

In my eyes it makes sense to streamline our product release schedule to quickly capture and defend these market segments. We do not need to develop entire new methodologies around each of these.
From a strategic perspective it seems like we need to differentiate between products with an extremely heavy engineering lift and those that can easily use existing methodologies.

We also need continued innovation and outside the box thinking - in the past few months we have seen incredibly detailed and well thought out proposals from @Matthew_Graham @Thomas_Hepner @Kiba @Lanks @MrMadila and many others. We should not be setting the barrier for Index release so high that these indexes never see the light of day. We should strive to encourage these kinds of submissions and to build a framework where smaller community led indexes can be launched and supported. Not all of these will be home runs - but some may become the next DPI or MVI.

INDEX.labs is the exact direction we need to be moving to support more product launches.

6 Likes

Hey @overanalyser we briefly touched on this the last time we talked about the Robot Index methodology (originally proposed as PMI), and I maintain that Index.Labs is a great idea which shouldn’t be held off by our engineering constraints.

Similar to the point raised by @DefiJesus & @Thomas_Hepner in this discussion as well as in this one, 2 months after the initial PMI proposal has gone in I am quite concerned about 2 things :

  • Not being able to push through the Robot Index concept at least up to DG1 vote and work team analysis. As an aspiring community methodologist, this would be all the more disappointing considering the significant amount of time & effort invested already or still being invested in this complex methodology (at the expense of other fields of contribution within the Coop).
  • Not being able to innovate and iterate fast enough on “real” product launches (not just prototypes) and carry a growing inertia while the market keeps developing at a quick pace, which might inevitably result in Index Coop not anticipating / missing fondamental trends.

There are solutions popping on a regular basis out there for community governed asset management which also reduce the engineering or regulatory overhead - I’m thinking for example about Enzyme Finance , Babylon Finance or Syndicate DAO with which the Coop could explore potential partnerships.

These are options that I am also exploring for the bloom .fi DAO and give reasonable space to move into, I’m happy to work on connecting the dots with these communities if a consensus is met around this.

1 Like

Strongly for. (Disclosure: I am 1/3 proposers of CWI)

DeFi would not exist today without someone, somewhere, at some point taking a risk and experimenting. Yes, the failure rate can be high but the net result is value creation. The biggest risk is taking no risk.

An incubator program for basic, easy-to-launch sets (on v2?) lets us experiment in a low-cost, low-risk way. As @BigSky7 said, we might just let loose the next DPI or MVI so the risk/reward from this perspective is skewed to the upside. Not to mention having ideas ready to go pre-cycle rather than mid-cycle (DATA index :eyes:). It also promotes creative initiatives that at the moment risk being stifled as ideas get bogged down on the forums rather than letting the market tell us what works and what doesn’t.

5 Likes

Dear all, following up on this and also on @Thomas_Hepner 's nice update about the DATA Economy Index, I strongly believe that outside of engineering or liquidity support one of the main benefits of Index.Labs would be on the business / growth hacking side of things.

In my mind this particularly applies to aspiring community methodologist who could benefit from the mentoring of the Coop’s established growth / marketing team as well as from a wider reach via the Coop’s (or Index.Labs specific ?) channels, for example to conduct market surveys.

This would be an efficient way for the Coop to coordinate the methodologists prospection efforts towards potential partners or investors, instead of relying exclusively on individual approaches :arrow_right:example : “Index.Labs currently has product A, B & C under incubation. Which product are you most excited about ?” …

3 Likes

Now we are talking my language.

And I like the way we are thinking. However, maybe we should zoom out and not limit Labs to launching funds. Labs could become our process for taking ideas to market for both internal tools and product offerings.

A process with specific steps for validating ideas prior to requiring engineering resources. This would be a program for gathering evidence of product/market fit and a system for prioritizing the projects we invest in.

Levels:

  • Sprints
  • Incubators
  • Accelerators

Fortunately, I spent the previous 4 years building a similar program for a large European bank. It included the methods, tools and training for their innovation program. I would love to implement something similar at Index Coop should we consider continuing this conversation.

4 Likes

Staying on top is all about speed and evolution, this is true of traditional organizations, physical technology, software, manufacturing, etc. As I read it, INDEX.labs would enable the Coop to add products at a greater pace, whilst allowing the engineering team time to focus on the more complex products & pressing issues. Ultimately this contributes towards Index Coop continuing to be the leading creator & provider of crypto index products.

Additionally (this is the part which excites me the most) The Coop increases its chances of stumbling upon something that works really well & drives big increases in AUM. No one can be sure what that ‘thing’ is, but if we cast the net wider we increase our chances of being in the right place at the right time.

A very interesting idea which I am supportive of

2 Likes