Introduction
Turing Meta exists to offer services to companies by translating cutting edge biology research into practical solutions to help solve the challenges of businesses and organisations. The history of much of this research is available online on www.turingmeta.org
The main result of this research relevant to businesses is a theory of ‘option farms’.
This technical whitepaper on option farms explains what they are in detail and briefly explains, via concrete examples, the singular modelling theory behind the wide range of offerings of consultancy and training from Turing Meta. Utilising this concept of option farms will enable your business to gain a competitive advantage. Further posts on the website www.turingmeta.org will highlight the business benefits of this modelling and also highlight the likely gaps in business organisation that option farm analytics, and training in option farm management can address.
You can read this article online or download as a PDF below:
Option Farms
An option farm is a different way to think of decision-making. Often decision-making is thought of this way: Different options are exclusive items we come up with and choose between, materialising just one option, among many considered, as the choice or plan to execute. In a company, at an executive level, it is very common for managers to present several options and ask the executive team to choose between them, normally with an option that is already recommended. In contrast, an option farm involves pursuing multiple options simultaneously to contribute to an objective and conserves the flexibility to switch options as things change. It involves making fewer big decisions but instead focuses on cultivation of as many options that can lead to success as possible.
Computational economics & decision-making
This method of behaving, involving acting in a way that keeps as many options open as possible turns out to be a powerful way of decision-making which has the potential to lead to the emergence of a huge variety of complex behaviour, as shown by the cutting edge research of biological computer scientist Wissner-Gross.
Turing Meta exists because of the basic insight that businesses and organisms are not that different. They both learn, grow and adapt, face the problems of uncertain conditions, and changing environments and ecosystems.
What the Wissner-Gross paper about biological systems and decision-making shows, is that when we think about what intelligent behaviour looks like, one way to think about it, is that it is about always keeping your options at any one time open. Entropy maximisation is a principle that Wissner-Gross has adapted to model this concept of keeping your options open with a physics-like concept that makes this precise (entropy is derived here from the physical concept but these details don’t matter).
Keeping your options open is a great way to deal with uncertainty, when you are leveraging the work of other people or agents whose contribution is unreliable or uncertain. This principle is effectively what the paper above explores, and it is probably best thought of as a constraint on how we act, rather than an objective.
But beyond dealing with uncertainty, the paper goes further and shows that keeping your options open can lead to the emergence of more sophisticated sequences of decisions that look more intelligent. This insight is significant when we consider business strategy relies on a sequence of decisions taken over an extended period of time often by different people and with varying objectives and conditions.
In one example, the paper shows that complex behaviour reminiscent of walking could emerge from this deeper principle of keeping one’s options open. There is a similarity between an ‘upright’ unstable pendulum which as a result is maximising (‘future’) entropy in the sense used by Wissner-Gross, and the similar biological behaviour of walking which is also an ‘unstable’ configuration which also increases entropy in the sense used by Wissner-Gross.
A modified version of thermodynamics causes a pendulum (green) swinging from a sliding pivot (red) to stabilize in an inverted, and normally unstable, configuration, from which it has a greater variety of options for its future motion. Credit: Physics Focus / A. Wissner-Gross/Harvard Univ. & MIT
The reason keeping your options open is so powerful
The reason this principle is so powerful is that for complex behaviour you often need an extended sequence of decision-making choices that add up to something greater than any individual choice. A simple example is the sequence of decisions that lead to the typing one by one of the letters that lead to a well written piece of useful program code. The same goes for a sequence of decisions taking over a period of time by an organisation. So to achieve something complex it helps that you need to retain as many options at each step going forward, so that the full range of options allows something good to happen later by still being able to make a sequence of choices from the full range of options at each step.
Credit: Dreamtime, Adam Timlett
It's like typing a line of program code that specifies some sophisticated action. Imagine that, as you type, the letters available on the keypad to continue specifying the program get fewer and fewer, until you only have a few letters to choose from near the end of the line you want to write. Then you are unlikely to be able to specify a viable complex line of code because the commands you want to write can't be written now as they require letters that are now unavailable. But if the all the letters remain available then you can actually write something complex over the whole sequence.
This Wissner-Gross model is one in which rather than making a small number of big complex decisions, you rely on the complexity of the decision emerging from a greater number of smaller decisions which achieve by focusing on just conserving your options going forward. Like typing a line of code from the full range of letters, such activity tends to lead to more complex behaviour than can be specified in a shorter line of code that could be written in one step. Hence, arguably this model of decision-making has the potential for greater sophistication in the resulting complexity and viability of the decisions, than a model that focuses on optimising for the 'best' choice at any given point in time.
Keeping your options open means making fewer big decisions
What Wissner-Gross shows is that it turns out that just paying attention to making decisions which keep the options open itself leads to, what looks like, valuable complex behaviour. So this leads to a different theory of decision-making from the one that focuses on the best single choice at a given moment in time. However, as explained in another Turing Meta article, there is a tendency for managers to focus on making a few big decisions, rather than lots of smaller decisions. Focusing more on option farming can address this bias.
Organic decision-making
The Wissner-Gross approach also addresses not having all the information at a given point in time. Making the 'optimal' decision at a given point in time necessitates that you are correct in the assumption that all the relevant information is available at that point in time when you make a big decision. But more 'organic' decision-making doesn't assume this vantage point, and accepts that other information and work by other agents available at a later point in time might be needed for a more optimal decision sequence. Therefore, we construct the more optimal decision organically by keeping options open, which allows later information and agency of other agents which comes in later to contribute to the sequence of decisions. This increases the odds of compatibility between earlier choices and later information and subsequently desired choices. That longer sequence of smaller decisions can then specify something more complex and viable between a set of agents making choices over a sequence of points in time than any big decision taking at one point in time.
Keeping options open is better for long-term results
Metrics which focus on optimising for a result in the shorter term, by the same token tend to involve making decisions which are less complex than might be required for a longer term result, as they tend to exclude options required for more complex longer term success. This is also why decision-making under constraints which conserve options also leads to longer term better results. An example of this is described by Bill Gates book ‘How to Avoid Climate Change Disaster’ in relation to aiming to reduce carbon emissions in the shorter term, rather than getting to zero in the longer term, as Bill points out, these are actually different things. Focusing on keeping your options open is another way of achieving the same result of longer term success but without needing to be specific (which might require a crystal ball) about which options will lead to long term success, and without even focusing explicitly on a complex long term objective, which may be hard to plan for.
Helping to manage complex work and de-risk
In addition to Wissner-Gross’s work it also makes sense to pursue many options simultaneously as a way of achieving an objective. If some of those options turn out to be more costly, or take more time or resource, on average your risk of not achieving your objective is mitigated by having other options that are also being pursued at the same time, which are independent of the affected options. Complex work is hard to predict, even simple programs like the installer of some software update cannot in general be reliably predicted in advance. That is why the progress bar on an installation program often seems non-sensical. But by having lots of options for progress, the unpredictability of individual options for progress matters less.
Making tacit knowledge explicit
We already have tacit knowledge that keeping our options open is good, allowing room in decision-making for others and our own future selves to add information in. But a scientific model of this which Wissner-Gross helps to provide makes the essential principles explicit and allows us to better understand and use this knowledge and apply it more confidently to a greater variety of scenarios and with more precision. It helps us to communicate the rationale for our 'hunches' and also to sharpen our intuition. Using Wissner-Gross and translating the parts of an option farm into metrics can help to make our tacit knowledge about options explicit and trackable.
The four major parts to an Option Farm metrics
1. Having more options:
When we think about what options we have, one of the main things we want to do is to maximise the number of options we can simultaneously pursue to achieve some objective.
When we count how many options we have, we can, and should, include various levels of materialisation of those options. We can think of options that we are pursuing simultaneously, that are fully materialised as similar to a portfolio.
We might have several investments at any one time, say in several different projects, that can help to meet a business objective. But we can also think of options that are not fully materialised as part of our option farm, such as research that is going on into the possiblity of new projects that haven’t been ‘green-lit’ yet, or different ways we might have to achieve the same result, which are under consideration. All of these less materialised options are also part of an option farm, in contrast to a typical idea of a portfolio.
These non-materialised options are part of our potential room for manouevre, the kind of principle that Wissner-Gross embodies. They increase the chance that we will be able to make a complex and sophisticated sequence of decisions, and to be able to make decisions that work in the longer term, and even subject to uncertainty in the environment or decisions of others.
For a technology strategy of a company, less materialised options we also count in a metric might be research about what technologies we have the potential to switch to, or potential additional features to purchase.
For a network of contacts, when counting our options we might include potential contacts connected to people we are already connected to directly.
If we consider the use case of managing a transport fleet, we might include options for additions to the fleet at a reasonable price, e.g. due to flexible contracts with car dealers.
For an investor in a high risk startup portfolio our options would include research on potential additions to the portfolio as well as the actual materialised investments.
For a football team more options means both the materialised options of the size of the squad of players but also transfer targets and promising players coming through the youth team system
2. Raising the quality of those options
While we do want to have as many options as possible, not all options are of equal quality. Very often it is cheaper to have options which are less materialised, than options which are materialised. Therefore, managing the quality of options in an option farm often means seeking minimal levels of materialisation and metrics to measure this
For business objectives we want our options for achieving the objectives which are not already fully materialised, to not be ‘grey’, i.e. uncertain in their quality. Ideally we want those options to be more materialised in terms of predicted results, to be viable, resourceable, etc.
For a technology strategy we want our less materialised options to at least be affordable, to be genuinely a good fit with the business objectives and existing technology stack
For a network we want contacts to ideally become more real relationships, or at least have a minimum level of engagement.
For a transport fleet we want our existing fleet to be of high quality, roadworthy, up to date, efficient in terms of fuel use.
For a startup portfolio we want our materialised options to be good companies, run by trustworthy and visionary founders, ideally with experience
For a football team we want our squad to be fit players, of good quality, with a good balance in options for each position.
3. Ensuring the diversity of those options
For an option farm to genuinely be ‘large’, in the sense that it provides protection from the risk that if some options fail they are covered by other options, we need not only for the options to be many, and the quality to be high, but also for the options to be sufficiently diverse.
Options might all be very similar, which tends to mean that they might all depend on the same component to work. This is risky, as it provides the illusion of lots of options without a real difference between them to spread the risk. Therefore, metrics to manage an option farm can’t just look at the bare number of options, and their quality, but must also seek to measure their diversity.
For business objectives this means that we actually want holistic objectives which cover a full range of things that are good for the business, and objectives framed in a way that are neutral about how to achieve those objectives, so that we can leverage the full range different potential options for achieving them. Then we want to measure that there are indeed options which are different ways to achieve those objectives that are relatively independent of each other.
For a technology strategy it means genuine diversity in a technology stack, e.g. by vendor, or by location of the compute, or by the functionality that it provides to achieve strategic objectives.
For a network it means a diversity of contacts that don’t all depend on a certain mediator or middle-man, and which come from a variety of social circles.
For a transport fleet it means options for transport that are non-trivial not just different colours of car, but different models and makes or a fleet powered by a vareity of fuel sources, such as electric, petrol, gas.
For a startup portfolio it means a portfolio that is diversified, so not all in the same industry or relying on same market trend or forecast.
For a football team it means not relying on one or two star players to be fit to have a chance of winning a game. It may also mean having the players to play different systems according to the opposition.
4. Favouring options with options – (Wissner-Gross - not all options are equal)
The final element of the metrics for tracking an option farm’s health, is to recognise that not all options are equal, because as Wissner-Gross demonstrates, it is very powerful to choose options which themselves lead to more options.
For business objectives this might mean metrics allowing us to choose options to achieve an objective which can fit with a variety of different levels of prioritisation as a project and still progress, or projects that can address a variety of objectives, so that if some objectives change, the same project can focus on the remaining objectives that still matter to the business.
For a technology strategy this might mean choosing technology and systems that are agnostic as to the stack they can fit to. It also might mean choosing technology that hits several business objectives, and technology that is function option rich, so can be expanded in a vareity of ways with multiple paths for developing the stack.
For a network this might mean metrics which help for focusing on contacts with the potential to have multiple potential types of relation status who might start as a colleague, but also become a customer, and also a broker to other contacts or social circles.
For a startup portfolio this might mean focusing on investing in companies that have the potential to make a significant number of pivots to focus on different markets or products if they have to.
For a football team, this means building a team which has many ways to win and which has players which can play in a variety of positions.
Summary
Option farms involves utilising metrics that mange the four major components to a healthy option farm:
1. The high number of options (at various levels of materialisation),
2. The high quality of options
3. The wide diversity of options
4. The choice of options with options
Depending on the actual domain that the option farm exists in, the way of creating these metrics will vary. (One example from a technology knowledge graph for a technology strategy option farm is discussed in this whitepaper by Turing Meta).
The competitive advantage summarised
Regardless of the domain, the power of option farms and a different way of thinking that they entail means that they and the right metrics have the potential to make you business more competitive, involving making decisions and sequences of decisions that are better for your company as a strategy in the long-term.
Option Farms enable your business to:
· Better de-risk strategic decision-making and
· Help your business to better cope with complexity and uncertainty,
· Increase its adaptive capacity and
· Allow complex sequences of intelligent decisions to occur over the longer term, and with minimal planning.
To find out more about how option farms can help you your business, or to explore other ways organic metrics can help with your business strategy, contact adam@turingmeta.org.
Download a PDF of this article below.