Brian Arthur and the Technology Evolutions

Years ago, while reading Stuart Kauffman’s At Home in the Universe, an essay on Self Organization and Complexity, I got struck by the following sentence:

“The car comes in and drives the horse out. When the horse goes, so does the smithy, the saddler, the stable, the harness shop, buggies, and in your West, our goes the Pony Express. But once cars are around, it makes sense to expand the oil industry, build gas stations dotted over the countryside, and pave the roads. Once the roads are paved, people start driving all over creation, so motels make sense. What with the speed, traffic lights, traffic cops, traffic courts, and the quiet bribe to get off your parking ticket make their way into the economy and our behavior patterns.”

This statement – as the author explained – did not belong to Kauffman himself but to W. Brian Arthur, economist, and very well known inventor of the El Farol Bar problem, a landmark within the Game Theory models.

I found – and still find – this Arthur’s statement enlightening: Economy as an ecosystem where major forces may change forever the overall landscape.

It also has a direct impact on Competitive Strategy, specifically when it comes to the possibility of Changing the Rule of the Games, like when transistors were used instead of tubes or Nintendo WII introduced an innovative remote control, which dramatically changed the way to think about video games and related user experience (just to make a couple of examples).

Since then, for many years I have been waiting for Brian Arthur’s masterpiece, the book that would have represented a sort of new start on how collective economy should have been conceived.
After 15 years, the book in subject has been finally published and titled “The Nature of Technology – What it is and how it evolves”.

The Author spends an entire book to define what Technology is and how it evolves and co evolves: the book per se is fine, but after 15 years my expectations were pretty much different. Personally, I was hoping to see some light shed to some fundamental questions like:

— If a new technology comes is, how does it spread within a Market?
— How long does it survive before changing into something else?
— When an ecosystem is a closed circle (a virtuous circle), how much robust is it to external disturbance?

Well, if you expect this kind of questions – and possibly answers – just forget the book.
If you are interested in dissect the technology definition, you might be interested to have a look at it.
As per myself, I’ll give Brian Arthur another chance and wait for another 15 years…

Less is More - The Steve Job's Way

There is a natural trend within most of the Companies – product based – which is so widespread to be considered as a Law of Nature.
At the beginning of its life, a “Forerunner” Company intercepts some “unexpressed Customer needs” which become “the Product”.
If the Product is successful in terms of functionalities, market positioning and marketing, then it gets bundled to those customers which concur to define its own “Customer’s base”.
Incumbent competitors may try to get the same Customer’s base entering the market with a similar product: when this happens, then the typical reaction of the Forerunner is to add new functionalities, to make its own product more rich and valuable. Differentiation – as a process – is practically what always happens and contains the gene of the Forerunner’s death: if the product gets more complex (e.g. improves/evolves, as the saying is) the Customer’s base shifts along as well, finally reducing its size, compromising the sales revenue and offering competitors the chance to fill again those initial unexpressed needs that were the original motivation to have “the Product” in the Market.
If it is not yet too late, a further reaction of the Forerunner is then to consider the product as a commodity and differentiate in Services, possibly compensating the product’s loss of value. But again, this form of differentiation will change the Customer’s base for some extent, moving the whole Company in an unexplored land.

The Stars Analogy
It is easy to see that what happens to product based Companies recalls the life of Stars.
A Young Sun grows in time becoming a Red Giant and finally falls into one of two possible conclusions: a Black Hole or a somehow eternal Pulsar, a White Dwarf.
Many companies easily get to the Black Hole condition, sinking without any remedy.
Others, niche based and offering high value services, may continue as Pulsars are used to.
But is this analogy just an analogy (and then with a possible different “Finale”) or a mandatory path?

The Steve Job's Way
I recently read this book “Leander Kahney - Inside Steve’s Brain”, whose content is focused on the strategy followed by Apple Computers, mainly along the guidance of Steve Jobs.
The most surprising element of Steve’s strategy is that “less is more”.
Apple's strategy is not focused on creating products which finally grow in complexity under the weight of new functionalities. On converse, the mission while conceiving a new success consists to add functionalities which simplify the use of the product through the interaction with the other functionalities already in place: a continuous positive feedback mechanism where any functionality may exist only if of some help to the others.
Any nice functionality which does not fulfill this approach gets discarded. Old functionalities which may be simplified with a new approach are removed as well.
This view sounds to me quite unique on the Market and allows to read the Star Analogy under a new interesting light.
It is certainly true that for most of the Companies the Star Analogy is a mandatory path but this is not the result of a “Lex Naturalis”: it is just the result of a bad Strategy.

Search Engines and Stephen Wolfram

Nowadays we all have experience of what a search engine is and how much useful it is. What is sometimes not that clear is the Strategy model they are based upon.

The main service they provide is the search for documents based on keywords. They all provide such a service, meaning Google, Yahoo, and the like, but with some differences.
Yahoo – when Google stepped in – was at its fullest, implementing the concept of “portal”, that is search for documents plus news, e-commerce, finance news, anything that could be brought at the attention of the user through this communication vehicle.
On converse, Google started offering a search engine that could be quicker and more reliable while providing results. What could really hit of Google was that the search engine home page was really bare and the first question “how do they make money out of it ?” could sound quite natural.
Then – after some time – something changed. The increase of speed and the greater accuracy of Google search results brought more advertising in, users started to know that advertising on Google could work better and took advantage of it. Advertising brought at the Google management attention that the more the content is present in Google – emails, blogs (like this one), Earth maps, even documents written directly through a software as a service model given by Google for free and the more effective the advertising could be (and the higher the revenue could be as well).
So, if it is still true that a bare home page is what we get typing www.google.com, it is also true that a universe of services built by Google itself or acquired through merge and acquisition actions – like YouTube – with the main purpose to bring always more content in – and then feed ads investment – are now available to users.
The Strategy pursued by Google is then simply this: feed the contents, search for contents, ads on contents.
Any of these three key processes sustain/fit with each other, implementing one of the cornerstones of Competitive Strategy. Looking at it we might also think that if operational effectiveness should be already in place, then the barrier for possible newcomers could become higher and higher as the time goes by…

So what Stephen Wolfram has to do with this?
Stephen Wolfram is a scientist – someone who got the PhD in theoretical physics at the Caltech at the age of 20 – and has contributed so far with many publications, many important results and – as an entrepreneur – being the founder and CEO of Wolfram Research, developing the software platform Mathematica.
This month Stephen Wolfram is going to launch the WolframAlpha.com computational knowledge engine.
The terms “computational knowledge” puts in evidence Wolfram’s scientific background so that – practically – the questions “what is it about?” and yet “how does he make money out of it” again come up quite naturally.
Apparently, it is another bare page which enables the search for documents but – as we can understand from outside – with two important differences:

1. Searches are made in English natural language, not just typing keywords
2. Answers are not lists of documents but a structured solution to the question posed by the user

I haven’t seen it in action so far, but the consequences of such a service may be easily spotted: the more final users will get accustomed to receive human like answers – supported by content, graphs and the like – the more the content will be provided by those who want to advertise their own interests in a way they could emerge as the preferred answer (or part of it). One thing is to get easy ads that may – or may not – get your attention, another is to get a “solution” where the ads are not just a possibility but the answer.
This kind of service will not just take advantage of content already available through other search engines, but will definitely be feeded by a “new kind of content” – more presentable and structured – and this new Company, as the critical mass gets reached, will design new ways to self sustain the service interest.
The Strategy pursued by WolframAlpha will probably be this (a variant of Google’s): search for any content, assemble the solution, ads on solutions and feed the solutions.
This sort of dynamic Wikipedia is going to change the rules of the game and this is always the worst possible risk within the Market arena when playing the Game of Competition.
Whenever Wolfram should make this Company public, just keep an eye on it…

Theoretical Strategies

In the previous article, the term Strategy has been defined – and refined – three times. Let's now see how we can make an effective use of it. The first definition, pretty academic and clearly referred to Porter’s classical statement has been followed by two concrete definitions, which underlie a methodology while facing with Strategy analysis:

Strategy = Strategy Design and Analysis + Strategy Implementation (a.k.a. Business Plan)
&
Strategy = Strategy Map + Business Plan

The development of a Business Plan is an activity which requires – on average – a huge effort and is substantially the financial proof that a Strategy might work. Without this insight, the Strategy Design and Analysis – possibly through a Strategy Map – may be considered as a Theoretical Strategy.

Are Theoretical Strategies useful ?
But in this case, which is the usefulness of defining a Strategy without any Business Plan? The answer to this question is quite profound and might be summarized in the following two rules:

1. If a Theoretical Strategy works, then the Strategy obtained adding the Business Plan might work
2. If a Theoretical Strategy does not work, then the Strategy designed along with the associated Business Plan will not work

The feasibility of a Theoretical Strategy is a necessary but not sufficient condition to have a Strategy really work.
Therefore these two statements make a Theoretical Strategy an interesting and useful tool, since it allows to provide an insight for any possible Company acting within a Market, foretelling whether the Company is inherently destined to fail or has good chances to play its game and succeed.
It also allows to define Theoretical Strategies that might be adopted within a specific industry depending on how far the various Companies – that are part of the examined industry – are from the designed Strategy.
A Theoretical Strategy is not a best practice: best practices bring different Companies to behave similarly while the objective pursued by designing and adopting a Strategy consists on being different by competitors.

The “best practice” approach has been the Trojan Horse used by Consulting Companies for a couple of decades, claiming that a best practice may reduce risks: unfortunately this brings all Companies in a vertical Market to converge on the same arena, competing with the same offering in the same way. As a clear example you may have a look at the way Telco Companies are substantially interchangeable…

The Essence of Strategy

The word Strategy, when applied to Market Analysis, risks to be used as a buzzword that means everything and nothing. Some people apply the term to define a generic approach of a Company within a Market; others think in terms of Company Strategy, Vision or even Mission as the same thing. Others again believe that Strategy and Business Case (or Business Model) are synonyms. A number of people even think that having read “The Art of War” of Sun Tzu is enough to know how to steer a Company.

A Definition of Strategy

In a short paper published by Michael Porter in 1996 (Michael Porter – What is Strategy? – 1996 Harvard Business Review ©), Porter’s definition of Strategy is articulated into three points:

1. "Strategy is the creation of a unique and valuable position, involving a different set of activities”
2. “Strategy requires you to make trade-offs in competing – to choose what not to do”
3. "Strategy involves creating fit among a Company’s activities”

The “creation of a unique and valuable position” is the objective a Strategy should always try to pursue. Specifically, this means that a Strategy is successful when the Company covers and owns an exclusive Market, where Clients – that may grow or not in time, depending on the Market size – are substantially not switchers.
How to create a unique and valuable position? In a way that Company’s activities – whatever the granularity chosen by an external observer – are reinforcing each other.
The idea of doing things differently is both a way to have something unique and distinctive and prevent others from imitating a successful set of activities.
Therefore, keeping the definition in one single statement, here is how the term Strategy may be defined:

“Strategy is the creation of a unique and valuable position, involving a different set of self reinforcing activities”

From Definition to Practice

The statement above may be used to design an effective Strategy on one side and understand whether or not a Company has an effective Strategy in place on the other.
Despite from the fact that a Strategy may be designed in an effective manner, later on it has to prove itself through the cash flow analysis.
For example, we might design an interesting high level Strategy for a hypothetical Railway Company and then – given the “as is” analysis of a real case – discover that moving from the “as is” to the expected “to be” is unaffordable in financial terms or has to be distributed over many years to make it happen.
To remark the distinction, the Strategy Analysis and Design is somehow “theoretical” while the Strategy Implementation is the so called Business Plan.

Strategy = Strategy Design and Analysis + Strategy Implementation (a.k.a. Business Plan)

The article of Michael Porter offers an interesting high level way to describe a Company’s Strategy Analysis that he calls Activity-System Map.

Fig.1 – Porter’s Mapping Activity Systems - 1996 Harvard Business Review ©

Such a map puts in evidence the set of activities (actually Business Processes, which are the sum of a set of human or IT activities) and a simple relationship among them.

Strategy Maps

There is anyway a better – and possibly more effective – manner to describe a Strategy Design and Analysis in terms of activities and relationships, which is known as Fuzzy Cognitive Map (or just Cognitive Map for simplicity) and comes from a set of ideas developed more than two decades ago by Bart Kosko.
This kind of map allows to add the following dynamics:

1. The enablement of Business Process “a” enables the Business Process “b” (and for symmetry, if “a” gets disabled, same thing happens to “b”) – to express this concept a “+” sign may be used
2. The enablement of Business Process “a” disables the Business Process “c” (again for symmetry, if “a” gets disabled, “c” gets enabled) – to express this concept a “-” sign may be used

These two simple rules allow simulating the evolution of a Strategy Map as a system and observing its dynamics, which may converge towards a stable position, oscillate among different states or never reach an asymptotic behavior. Initial conditions may be set as desired as well and some states might be forced for some time to maintain a fixed status and then released all of a sudden to see in which behavior the system would land.

Fig.2 – In the example above, Initial Funding drives Marketing campaigns which pushes for Product Popularity. The latter makes Revenue grow and they keep Marketing alive. At the same time, the more the Marketing is active, the more is the Competitive Pressure which – on converse – decreases Product Popularity activating a reverse cycle (i.e. less Product Popularity implies less Revenue then less Marketing etc). This toy example represents at a high level the way Mobile Phone operators usually compete within the Market.

This kind of map is very useful to analyze Strategy robustness and stress it under specific Market conditions.
As it always happens with simulations, if the outcome should be a “yes, it works”, this might be taken as a “maybe” in real world, as the Business Plan filter should be successfully applied as well.
But if the outcome of the simulation should be a “no way”, then we could be sure that in the real marketplace the result would be the same.

The “equation” introduced before becomes the following:

Strategy = Strategy Map + Business Plan

In Summary

Designing a robust “real world” Strategy is an iterative two steps activity: the Strategy Map effectiveness first and the Business Plan feasibility afterwards. The Strategy Map applied to the “as is” condition of a real Company drives the Business Plan; Business Plan constraints allow going back to the Strategy Map and refine/optimize it.
The result of this “self reinforcing” iterative approach is itself an overall Strategy that follows the definition introduced in the first paragraph.

Preys, Preyers and Market Behaviors

A common pitfall when talking about Business Strategy is that Market size is somehow undefined and then perceived as unlimited.
If this baseline, under a certain approximation, may be seen as true thinking about the final consumer – since in this case the number of potential Clients is incredibly higher when compared to the number of Companies that might satisfy their needs – this is not the case when looking at specific market segmentations involving either final consumers or intermediate entities, like other Companies.

In particular, Market size might be static for some time and then considered limited or growing steadily – typically linearly – or through a step which doubles or more the Market from a certain moment on.

The scenarios that a Strategy Analyst might face with are then the following:

1. One or more Companies competing for a static Market
2. One or more Companies competing for a Market which is growing at a certain rate
3. One or more Companies fighting for a Market that under certain conditions – e.g. an innovation related to the product being sold or a sudden availability of cash – immediately grows in size all of a sudden

The question that immediately comes up is: how does the Market behaves in such scenarios?
In other words: is this behavior predictable?
The answer is yes and may be given studying the Prey vs Preyers equation in the version known as Lotka-Volterra model

dx/dt = ax (1 – x)

that as a discrete differential equation becomes

x(t+1) = x(t) + ax(t) [1 – x(t)]

This model - where x represents a Company or an Industry and a is a growth rate - may be applied to examine a single Competitor – a Monopolist – who acts within its own Market or two/many Competitors who compete with each other to win the Market.
Results – already mathematically explored in literature – are quite interesting.

Monopolist and Market Size
When a Company lives in a Market thanks to a specific competitive advantage – whatever the reasons that may be the source for such an advantage – it may substantially grow in one of three ways.
Slowly – and in this case it will reach the asymptote of the Market quota after some time. This kind of Company experiences a good rate of growth along its childhood and then reaches the fullest while aging. The adjective that may be used to describe this condition is “stable”, meaning that such a Company – in a world where innovations or new entrants or suppliers are not a threat – is not a risk for itself. The interesting point is that the growth of the Company is a consequence of the rate growth but the curve which describes the growth – a reversed exponential – has always the same shape when the growth rate is under a certain value.


Fig.1 - Slow growth

Quickly – this is the case of a Company which wants everything now, maybe to establish a consolidated position before others might step in or just because it is necessary to grow quickly given a business plan – and a financing plan – that requires this approach to justify a return on the investment. As wisely observed by Michael Porter in his article “What is Strategy”, growth – or exceptional growth – tends to drive revenue reduction. A Company which grows very quickly “consumes” the Market liquidity at a high speed, then a period of low sales shows up and a resize of the Company or a strict cost reduction policy is necessary to stay in the Market waiting for the next wave. This kind of Company experiences some ups and downs whose reasons are usually searched within many factors – Management, Strategy, Focus, Market mood – but the ultimate driver is the excessive growth that wants to be pursued.


Fig.2 - Quick growth

Extremely Quick – and in this case a disaster immediately shows up. A Company whose growth is exceptional and is able to saturate the Market in the short term builds an infrastructure – an organism – that immediately afterwards cannot find cash available enough to live or survive. Despite the mathematical model in this case behaves chaotically – giving the impression that it is always possible to hook at the next positive wave and then resort – the truth is that the Company dies. The size of the Company is not the reason of its death – this happen whatever the size of the Company: the unique reason is the exceptional growth rate which exhausts the cash availability – i.e. the overall spending capacity – of the Market in which it operates.


Fig.3 - Extremely Quick growth

Many Competitors in the Same Market
Given the examples above, it should be clear which might be the behavior of a Market where many Competitors are fighting for the same source of revenue – i.e. within the same “Industry”.
If on average they exhibit a low attitude to growth, they will all get their own position within the available Market. If they should grow quickly or too quickly – or even if one of them should be able to perform this way – the survival of all of them might compromised and an economical catastrophe would emerge.
The latter consideration is particularly helpful to understand those exceptional growths – both on promising (e.g. the new economy) – and on consolidated Markets (e.g. Banking, Finance, Real Estate) that ends up with a general crisis. The system – as a whole – burns so much liquidity that the sustainability of the systems itself is then compromised.
It helps – strategically thinking – to understand why a Company who wants to enter within a Market through a diversification growth approach might decide to make the “price war” to burn the Market trying to establish itself as the main leader after some time.
It is also necessary to understand why – under certain conditions – a free market with no rules may lead to unpleasant conclusions. Years ago the China government had to fight against internal free market desire, offering the explanation – among others – that a sudden freedom would have led to chaos. Despite the means adopted that time and the unfortunate events the public opinion had to witness – like the Tiananmen protests – the Lotka-Volterra model clearly proves that their fears would have possibly become true.

What if the Market Size Grows?
Market size is the consequence of many factors: the number of Clients available – which might be static by nature (e.g. the number of Banks on a certain period) – or the product positioning: the higher the price, the lower the number of Clients (this will be a subject for another article on this Blog).
Sometimes a new technology or a product improvement makes the Market size grows: in this case, which is the behavior that we might expect?
It all actually depends – again – on the growth rate: if the Market growth rate is higher than that of the Competitors who are fighting for it, then a period of stability is granted. Nevertheless, when the Market size has reached its new maximum size or whenever the Competitors aggressiveness should be higher, then the behaviors depicted before would immediately show up. A Market growth shifts a possible problem in time but does not prevent it from happening.