Are you a developer?

Join my startup: Elody

Work together with other developers and build a Gestalt AI.

Elody can make your programs accessible to
non-technical users.

Feedback Funding

I was thinking about the best way to get funding for my startup and none of the existing systems were a perfect fit. So I invented my own.

This is a financial instrument that is specifically designed for crowdfunding fast-growing tech startups more effectively than existing types of investment.

Feedback has been overwhelmingly positive, but nobody is sure about the legality of it all (not even the German government agencies I asked).

Unfortunately, actually implementing this system would be a lot of work for me and I am still busy with my startup itself. I also have no idea about the legalities involved in creating a new system for funding a startup.

Hopefully I can convince someone else to implement the system, then sign up with them as a customer. It would be very easy for an established bank or crowdfunding platform to offer this service. Perhaps the whole contract could even be fully automated using a cryptocurrency?


This financial instrument has a very different risk/reward system than traditional crowdfunding methods: Instead of a large chance to lose all your money and a small chance to make it big, FeedbackFunding gives you a large chance to make a fixed amount of money and only a small chance to lose.

This is because you do not bet on "how big is this startup going to become?" but on "will this startup keep growing?" You don't need to predict what will happen in ten years, only what will happen over the next few months.

I call it FeedbackFunding because it is a way to get funding in a continuous fashion, so that the startup can use earlier investments to improve itself and thereby convince people to invest more, in a feedback loop. While this basic idea is also present in startups that go through multiple funding rounds, it is much more important for FeedbackFunding.

What problems does this financial instrument aim to solve?

  • Startups find it very hard to get initial funding.

    When they do get funding, they get much less money per share for earlier investments than for later ones. This means that they would have received significantly more funding in exchange for the stock they gave away if they had traded all of their stock at the same rates as in the last round of investment. There is (currently) no way around this, since the startup could not survive to reach the later rounds of funding without the investments made possible by the earlier rounds.

  • Crowdfunding a startup often lacks immediate incentives for investors.

    Investors of crowdfunded startups often get only a token reward for their investment that is not worth the money they invest. If they do get a small share of the stock, it will take years before they start seeing a return on their investment. While such long-term investments are the bread-and-butter of professional investors, crowd-funding investors are mostly just excited by the startup and its product itself. They don't care about the long-term gain as much as they care about the idea of supporting something they like. This is why contemporary crowdfunding platforms often reward contributors with thematically appropriate tokens of esteem, rather than with money. If the contributor of a crowdfunded startup were able to get a small return on their investment after only a short time rather than after several years, this early reward would motivate them more strongly than the hope for a large payoff in the distant future. A token of esteem is perceived as more rewarding than a promise of money in the distant future, but a promise of money in just a few short months is better still.

This is a financial scheme that aims to solve both problems. It can also be used for other products, but its most immediately apparent use is for funding startups, so these are used as an example.


This financial instrument is designed for a specific type of asset:

  • The value of the asset lies in the long term. It is currently worth very little, but has massive potential for growth.

  • The value of the asset is unlikely to be either stable or to fluctuate wildly like stock of publicly traded companies does: Instead, its value is likely to continue increasing at an exponential rate until it either plateaus or drops off sharply.

This is an excellent fit for the stock of many tech-startups, especially ones based around social networks.

How does it work?

The startup creates an instance of the financial instrument [A]. In practice, many such instances may be made available simultaneously, and they may be bought in bulk for convenience. This does not change the way this works. Each [A] may be considered separately to calculate earnings.

Initially, each [A] is sold at a price determined by the startup. Let's call the starting price [I]. Additionally, the startup defines parameters for the [A]: the object to be sold [O] (for this example, this will be a set amount of stock in the startup), the end date [T], the step factor [S] and the payout factor [P]. [S] and [P] should be greater than 1 and [S] greater than [P]. Each [A] also has a current price [C], which initially is [I].

For purposes of clarity, assume the following parameter values as an example:

  • [O] = 0.1% of the startup's stock
  • [I] = 10 USD
  • [T] = 3 years from the publication date
  • [S] = 2.0
  • [P] = 1.2

The initial buyer of [A] does not get the object [O] that is actually to be sold. Instead, ownership of [A] will change hands continuously until time [T]. When [T] is reached, [A] is dissolved and ownership of [O] is transferred to the current owner of [A]. The trick to the whole thing is the rule by which ownership changes:

Anyone is able to purchase ownership of [A] at any time. The current owner of [A] is unable to refuse this, but is rewarded when this happens. Ownership of [A] may only be bought up at a price of exactly \([C] * [S]\) (the current price multiplied with the step factor). Part of this money is payed to the previous owner of [A], the rest goes to the startup. The amount received by the previous owner is \([C] * [P]\).

This means that if you purchase [A] but then someone else buys it off you again later, you are guaranteed a return on investment of \(([P] - 1)\), or 20% in this example. Consequently, the startup gains \([C] * ([S] - [P])\), or 80% in this example. After the [A] changes ownership, its current price [C] is updated to the new value. This means that the cost [C] of [A] increases exponentially with factor [S] as [A] repeatedly changes ownership.

The investor who finally ends up owning [A] when time [T] is reached is rewarded with the object [O], which in our example is 0.1% of the startup's stock. Whether or not that means that the investor made a return on his investment depends on whether or not they correctly estimated the worth of [C].

Any previous investors always benefit, because they made an ROI of \(([P] - 1)\).

Notably, they make their \(([P] - 1)\) ROI as soon as someone else buys the [A] off them, and can then invest their money again if they want.

With a normal investment, you gain an unknown ROI after a fixed amount of time. With this system, you gain a fixed ROI after an unknown amount of time and can simply reinvest if you want to gain a higher total ROI.

Finally, the startup benefits because they gain a continuous stream of funding that starts low and increases over time, without having to give out any additional amount of stock.

Special cases:

  • An investor may purchase an [A], then lose it to someone else (earning money in the process), then purchase it again later. This does not have any negative impact, and is in fact likely to be a positive motivational factor for investors: They already made money the first time they bought [A], and the company is still growing, so they are likely to earn even more money if they buy [A] a second time.

  • The system as described above can be exploited, so a minor change is recommended, though not strictly necessary:

    If two investors cooperate and take turns buying an [A] from each other, both keep paying each other back some of the money they paid earlier. This is not really a problem for anyone, but it could lead to some unnecessarily strange investor behavior.

    This can be fixed by adding another possibility:

    Even though [A] can only be bought at the fixed price [C] * [S], it is possible for the current owner of [A] to increase the price [C] further, either to make it harder for rival investors to take [A] from him, or to increase the profit he will make when they do.

    The trick here is that increasing [C] for an [A] that you currently own from A to B is actually cheaper than (B - A). The degree by which it is cheaper is just the degree necessary to make the above mentioned cooperation between two investors unnecessary:

    If [C] is only increased in steps of factor [S], so that \(\log_{[S]}(B/A)\) is a natural number, the formula for the money siphoned off by two cooperating investors is:

    \[A * ([P] - 1) * \sum_{i=1}^{log_{[S]}(B/A)} [S]^{i-1}\]

    This is equal to:

    \[(B - A) * ([P] - 1) / ([S] - 1)\]

    There does not appear to be a reason not to also use this formula when \(\log_{[S]} (B/A)\) is not a natural number.

    Therefore, anyone who already owns an [A] at a price [C] of A should be allowed to raise it to any higher price B not by paying the difference \((B - A)\), but by only paying

    \[(B - A) * (1 - ([P] - 1) / ([S] - 1)) = (B - A) * ([S] - [P]) / ([S] - 1)\]

    In our example, this is \((B - A) * 0.8\)

    Using this discount ensures that the investors have proper incentives, assuming an efficient market:

    The optimum behavior for an investor is to predict the true value of [O] at time [T], then buy any [A] with a cost below that value and raise its price to no less than \((1 / [S])\) of this value. If the investor's estimate of [A]'s value is accurate, nobody else will buy [A] from him, as the price of doing so would be above the true value of [A]. The investor will thus end up owning [O] for \((1 / [P] / [S])\) of its value in the best case. If on the other hand the investor underestimates the value of [A] and another investor is willing to pay [S] times as much, they still end up with the fixed ROI of \(([P] - 1)\).

Summary of earnings

  • The investor who owns [A] when time [T] is reached gains [O]
  • All previous investors of [A] gain an ROI of at least ([P] - 1). If someone bought [A] and then raised its price using the discount formula mentioned above, they have an even greater ROI. Incidentally, this is a useful motivational tool the startup can use to get investors to invest more money immediately, rather than waiting and only investing when they are more certain.
  • The startup earns 100% of the initial sale price of [A], and (([S] - [P]) / ([S] - 1)) of the remainder, which is 80% in the example. While this is of course worse than keeping 100% of the funding, this is actually an advantage that FeedbackFunding has over traditional methods, as we will see in the comparison below.

The only way for anyone to lose is if the startup fails, or if it unexpectedly stagnates while you are the current owner of an [A] and were counting on a continued growth. This is the same failure mode as in traditional startup investments.

The key difference between this system and a traditional system is this:

  • Traditional investments offer high uncertainty about getting a return on your investment at all, but have an unbounded potential ROI.
    Most people lose, but some people win a lot.
  • FeedbackFunding offers a fixed ROI for most investors, but has a much higher chance of success.
    Almost everyone wins, but not by very much.

It is also worth pointing out that the funding arrives continuously, as investors outbid each other all the time, and no special actions need to be taken by the startup. In particular, the startup does not need to approach incubators, accelerators and the like, which usually takes a lot of time and has a low chance of success. Perhaps most importantly, the startup only needs to offer its stock once, whereas a traditional approach involving multiple rounds of investment costs the startup a cut of its stock every time.

Comparison to traditional funding of startups

An example (this is not realistic, but shows the general idea):

A startup starts out without any external funding and needs to secure money so that it can grow. The startup currently has a lack of funding and no new employees can be hired for now, even though they would greatly help to increase profits.

Scenario 1 (traditional funding)

While the business flounders from a lack of new hires, the founders of the startup spend weeks of their time preparing a presentation for an Accelerator, Incubator, Angel Investor, or some other kind of traditional investor. After applying and presenting at dozens of them, they finally get an offer: 30% of their company for 10 million dollar. Glad for the search to finally be over, the founders accept.

Had they continued the search, they would have gotten a much better offer a week later, but they will never know. Now that they finally have money, the startup immediately hires a lot of people. The new hires are inefficient because they have all joined at the same time and there is not enough time to properly bring them all on board.

A year later, the money runs out again and the scene repeats itself as the founders go looking for another investor. When they find one, this new investor offers 100 million and also demands 30% of the stock. Now, the founders no longer control the company, but they don't have an alternative.

Scenario 2 (funding with the scheme described here)

Shortly after founding the company, the founders offer 20% of their stock, split into 2.000 [A]s, with [T] due in 3 years and [S] = 2.0 and [P] = 1.2, at a very low starting price. Through word of mouth, they convince a few hundred people of the worth of their company, purely as a side effect of their work, without spending time on courting investors.

Seeing the low current price of the [A]s and expecting the startup to perform well, some of these people immediately invest small amounts in the startup by buying up [A]s. The startup now immediately has a small amount of initial funding and can begin to hire employees. There is no period of financial dearth, nor do the founders have to spend time on wooing investors. They can concentrate on making an awesome product.

As the product improves and the company grows, people start becoming more and more convinced that the startup has a big future. The prices of [A]s rise gradually as people start investing more in the company. This leads to a steady stream of additional funding, allowing the founders to hire additional employees at a healthy rate.

The founders have still only offered the initial 20% of their company, and did not have to negotiate with any major investors. If they find that it is worth doing so, they may decide to offer additional parts of the company as new [A]s to generate more funding and ensure a faster growth. They can do this with exactly as much stock as they wish, and they do not risk losing the majority over their company unless they deliberately decide to offer more than 50% of the company's shares.


Even though the founders in the second case get only 80% of the funding due to the values of [S] and [P] they chose, they still end up with more money than in scenario 1.

After all, the founders in scenario 1 lost their first 30% for 10 million and their second 30% for 100 million, for a total of 60% for 110 million.

Meanwhile, in scenario 2, the value of each [A] gets updated and corrected upwards as higher bids come in and the [A] changes ownership. If the first investor from scenario 1 were to invest in scenario 2, he would receive his fixed ROI of (1 - [P]) = 20% after only a few months, while the startup would continue to gain money from those [A]s. This investor would gain less in scenario 2 than in scenario 1, but given the rapid growth of the company, he would receive his 20% ROI after only a few months, which is not bad at all.

In scenario 1, earlier investments are very inefficient compared to later ones, so the average efficiency goes down over time. In scenario 2 on the other hand, the efficiency of every investment is always fixed at 80%.

Summary of pros and cons


  • Investors start investing much earlier and even private individuals may start investing low amounts. It's a kind of crowdfunding.
  • Early investors make a profit as soon as new investors come along that evaluate the company more highly. In this way, people are rewarded for correctly predicting that a company is more valuable than it seems.
  • The last investors benefit in the same way investors usually do: they get stock in the company.
  • The company only has to give away shares once. There is no need for multiple rounds of funding, so no value is lost to earlier rounds where stock is sold at a lower amount of money per share.
  • The company does not have to spend valuable time and energy creating pitches for individual investors. Make the investors come to you, instead. If you can't attract high-profile investors, crowdfunding through private individuals will still work.
  • Assuming an efficient market and an informed public, the estimated value of the company can be tracked through the current evaluation of an [A].


  • The owner of the startup does not get all of the money the investors invest. This is offset by the fact that the company only gives away shares once.
  • Early investors who pay a low amount will only gain a low amount, which makes them less likely to want to invest. Instead of luring early investors with the potential for massive payouts, early investors are lured with the very high probability of getting a small payout very soon.
  • If the factor for the minimum required next amount of money is too high, the price of an [A] may get stuck quite a bit below the fair market price, because the next step size is above the fair market price, and no one is willing to pay that. This will only occur very late, once the startup is no longer growing. It's also not really a problem so much as it is a limit on the advantage provided by this scheme, since other financing models do not have this feature in the first place.

Optional features and additional musings

  • Optionally, the startup itself may be allowed to buy back an [A] under the same terms as any other investor, and to terminate an [A] it owns. If this happens, it effectively means that every single investor in the startup earned the same fixed ROI on their investment. Despite this, the startup still comes out ahead since the investments happened continuously over time.

    Under a certain point of view, this effectively turns the [A]s into a kind of loan from the investors to the startup, but where a traditional loan has a fixed rate per year, this has a fixed, predetermined rate / ROI but an unknown amount of time until it is payed back.

  • Instead of offering a single instance of [A] for a large fraction of the company, it may be more effective to offer a very large number of instances of [A] that are each worth only a small fraction of the company, and allow people to buy them in bulk. A buyer would specify either the total number of [A] instances they want to buy or the total amount of money they want to spend, and the bank (or the algorithm, if it's offered through a cryptocurrency contract) would make sure that the instances of [A] are chosen for sale in ascending order of their current price [C].

  • It should be possible for the startup to alter the parameters [S], [P] and maybe even [T] when an [A] has already been released (the initial sale has been made) and is being traded. The right to do so must be preserved contractually, as unexpected changes of these parameters could be harmful to investors.