Transformation of Commercial Networks: Unlocking $2T in Value

 18 Feb 2016

I’m in a network state of mind…  We are in the midst of a massive economic transformation and I can’t quite put my finger on it.  What influences how consumers and businesses are organized? What is changing? Who creates value? What new domains, networks and markets are being created? Where is margin flowing to and from? The hypothesis from Paul Graham’s Refragmentation blog has been keeping me awake at nights.

“that all these trends are instances of the same phenomenon. And moreover, that the cause is not some force that’s pulling us apart, but rather the erosion of forces that had been pushing us together.” Refragmentation

It’s as if the gravitational constant (the big G) is changing and new forces are driving formation of new networks, influenced by a rapidly evolving world of “weak links”. Information intensity has moved beyond “tweaking” 100 yr old business models to transform the design of industries, communities and people. I outlined several of these dynamics in Internet 3.0: Collaboration in Commerce.

The economists have been far ahead of us here, as have others. My reading this month:


I’ve truly enjoyed putting this blog together, as it is the basis for my company Commerce Signals. We are in the midst of a massive transformation of business and networks. Little has changed in Retail, Banking and Commerce in last 150 years beyond Scale Wins. Consumer behavior and new networks are disrupting traditional economies of scale (asset intensive) as well as new information economies of scale.

Effective networks are wonders of business and social interaction that largely re-inforce an existing pattern, product, or social structure. Networks are resilient to as they create value to all those connected… and this value expands as the network grows (network effects). The reinforcing nature of networks has proven effective in insulating participants from being impacted by change and keeping disruptive forces at bay.  Profitable companies are seldomly drawn to models that circumvent them or operate at a different margin/scale (ex innovators dilemma). New networks have reshaped how every entity can both consume and create services; thus resetting the forces that shape the design of a company (ie outsourcing/specialists).

Value in a network is created at the intersection thus value of the network = f( number and type of participants, # of services ) which roughly correlates to the number of intersections. Most of today’s networks focus narrowly on solving a specific problem for a group of similarly situated participants (ex Mastercard, eBay, AT&T, NYSE, …) and creating substantive standards for the participants to follow. The most valuable company in the world (Google) has created an open democratic network of heterogeneous nodes and services that have connected consumers and businesses in new ways. Thus demonstrating that there is more value created in connecting heterogeneous participants than peers. There are 1000s of companies with better data than Google, but with none with overarching network or rules for data exchange (ie principally focused on privacy and control concerns,  not the “how” of  technology transfers of data).  This is the problem my company is focused on.

Traditionally, business networks are UNICORNS. After all, competitors rarely agree on any standard to interact unless there is clear business incentives that lift all (big players) equally (think NASDAQ/Visa), or to the best of their ability. In this traditional world, economies of scale (assets and information) have kept profitability with large participants who themselves have their own networks of suppliers and standards. We are about to see many more Unicorns, as new business networks form that have far greater value (ie Google) from intersections of dis-similar nodes. In this future world, value is created by the orchestrator (think Amazon, Uber, AirBnb,..).

In the blog below I lay out a few of my hypotheses of networks.  These hypotheses are based upon my time with top platforms, networks, banks, retailers and technology providers. I don’t have the academic credentials … so sorry for the errors or any references I’ve unintentionally missed. Your feedback is appreciated.

Tom’s Network Hypotheses

  1. Commercial Networks reinforce a value proposition and are resilient to change. The economic value within networks is migrating away from nodes and shifting to orchestration.
  2. The winning business model over the last 100 years has been scale (asset and data intensity). New networks will diminish Nodal Scale Advantages. Data democratization will enable competition by smaller nodes operating in smaller agile networks at a much faster pace (Navy Seal teams vs. standing armies).
  3. These two models of Networks will struggle for dominance in next 10 years. First is Large Node clusters where established businesses engage in Structured Collaboration to create value with non-competitors (ex big bank competitors in a payment system network creating value for themselves—and others—at the nodes between consumers and merchants). Second is Orchestrator led Clusters of small nodes (ex Uber).
  4. New Clusters form on “intersections of value” (ex Uber) with primary margin capture by the orchestrator of the interaction (ie the center of the star network).
  5. Orchestrator-led Clusters (small star networks like Uber) are now the most effective network constructs for value creation and specialization. This type of  Clusters also allow for participant investment, discovery, trust and value exchange mechanisms the way the participant wants to play, not the way the large structure works.
  6. Networks of heterogeneous participants (ex Google) have greater value potential than that of homogeneous participants (eBay). Logically if value is unlocked at the intersection of participants, the intersections of dissimilar entities have greater value than that of peers. (This is our focus at Commerce Signals).
  7. Fortune 100 (Scale Nodes) will attempt to counter this small network cluster dynamic through structured collaboration with non-competitors, thereby realizing new value intersections. Example: to compete against Google.
  8. Consumer behavior and connectivity have enabled many new intermediaries that have disintermediated most traditional business models. Consumer trust, value and experience are keys to new network creation. Trust is domain specific, thus current networks are constrained by BOTH network capture and by trust extension.
  9. Open Networks allow for transparency, rapid evolution and scale… but not trust. thus few existing businesses seek this design.. thus are best suited for situations where no value is exchanged (ie Wikipedia) unless trust and authentication is embedded (ie Blockchain)
  10. Blockchain – If it were able to grow freely… it could disrupt everything.

I will comment on a few of these below..

Why do networks exist?distributed

Networks form in nature, business and society, from the atomic to interstellar. The reasons networks form are varied; thus it is worth assessing the shaping forces, and the processes that reinforce their growth.  Why do human networks form?

Humans are cooperative animals (Ridley, 1998). Consequently, our brain has developed to keep an inventory of our contacts (Dunbar, 1998). Connection rules have become essential for our survival. Social surveys show that we get used to circles of 5, 15, 35, 80 and 150 people.  However, we tend to meet more and more people and our cognitive abilities are not prepared for this. In a modern megalopolis, we feel lost. The expanding world has become alienating. Our only escape is to continually redefine and segregate a small world from the contact wilderness outside. Not only does small-worldness provide a key for efficient network search, as well as respecting our cognitive limits, but it is probably also a prerequisite for preserving our safety in an alienating modern world. – Peter Csermely. Weak Links pp. 315-319

We network as a method to interact (ie manage our efforts) within the environment (socially and in business). Networks reinforce value and enable trust.  Connectivity and free flow of information has transformed how humans network. Sociologically, we interact with a counterparty based upon the level of trust we have with them, the objective of the conversation, and the value we intend to unlock. In turn, we are restricted by the networks in which we operate. We may not (and likely don’t) know all the people who are best suited to help us is a question we can not answer.  Rather than being able to easily use a discovery function amongst all possible candidates with an answer, instead we are constrained to the nodes in which we can access and the incomplete answers our networks can provide.

Cycle times to discover and participate in many networks have transformed how we operate. We are no longer constrained by our geography. The efficiencies and experiences possible in new networks are displayed in both consumption and creation: Uber, Wikipedia, MOOCs, mail, Twitter, Facebook, phone, search, blogs, etc. In this hyper-democratized environment, we don’t care about race or religion when consuming services or ideas.  We care about reputation (creating and maintaining),  value, experience, speed, etc.

Democratization, transparency, and trust are HIGHLY disruptive to established businesses. How many commercial networks can you name where 100s of participants have invested billions of dollars? Businesses have benefited from operating in semi-closed network (e.g. ChasePay, Railroads, AT&T, United Airlines). Owning distribution and rules protects unprofitable products. When you own the hardware with the round holes, you don’t want to see square pegs coming at you—you try to shave them down and force the pegs into the holes you already have.  However, the forces of new networks (ie Google, Amazon, FB) and trust domains (Uber)has allowed new competitors to significantly disrupt supply chains I tell my kids we are transitioning from a world where everyone had their own patterns for “Lego-like connectivity” to one where each business component must seek to interoperate.

Consumer Trust

If I had a graphic artist and 2 analysts I would draw 3 network diagrams: one each for 1900, 1990, 2016. For each year I would attempt to graph the commerce and social network of a small city’s population. Remember the old adage “its not WHAT you know but WHO you know”. Reputations and Trust took a lifetime to build, economies of scale and asset intensity dominated high margin growth. My label for this project would be visualizing our social fabric. Ad Week on Trust

My guess is that we would see a tremendous shift in economic center away from local merchants and friends into broader value focused networks as swarms.swarm

New forces of Attraction

Per the Paul Graham Refragmentation blog, are new networks forming because of new forces of attraction or erosion of existing forces? My Answer: Yes. The great thing about networks is that they are “RESILIANT TO CHANGE”—the juggernaut continues (more or less) regardless of what’s going on in the world around it.  It’s the oak that can lose some leaves and branches in the wind, but manages to hang on despite all that.   But this same force operates to their determent. What is the leading success factor in Commerce and Banking in last 100 years? it is the success of “SCALE”. While the benefits of network effects are clearly understood, the downsides of “network capture” are less so. Competitors, which commit resources toward differentiating similar products in a common network  are less likely to create new products or find new outlets for those different products.

Payments Example – Visa and Mastercard are ENORMOUSLY EFFICIENT Networks. Payments could certainly be performed in another channel, and in a more logical fashion,  but there is NO MODEL that provides a better acceptance rate with merchants and consumers. Thus “magic” of the V/MA model is profitability of payments sits with issuers, and they in turn invest billions of dollars in using the networks and their rules.  Bank profitability comes from both interchange and other fees, but especially in the lines of credit (interest and fees) that come with the account.  Bank unsecured revolving lending has thus been “captured” by this efficient payments network. Issuer-led payments innovation can be constrained by the channel’s lending legacy. Thus the logic of Network Led Orchestration and Value Creation.

Hypothesis #1: Commercial Networks reinforce a single type of value proposition and are Resilient to Change. The economic value within networks is migrating away from nodes and shifting to orchestration.

I’ve written several blogs on this one and won’t expand (see Google Platform, Distributed InnovationApple’s Opportunity).

Hypothesis #2: The winning business model over the last 100 years has been scale (asset and data intensity). New networks will diminish Nodal Scale Advantages and data democratization will enable competition by smaller nodes operating at in smaller networks at a much faster pace (Navy Seal teams vs. standing armies).

Over the last 50 years top retailers and banks enjoyed economies of scale in data, distribution, and marketing/brand. The profitability in banking and retail was NOT in the network, it was in selling commodity products. What happens when the products, along with trust and reputation, can be distributed and monetized? When information intensity replaces distribution and asset intensity?  Look at Google’s shopping express for example, Google Express will ship goods to your door for free (twice a day). Amazon is offering same-day delivery on many cities as well.   How does a retailer compete with that!?

The other fantastic example is Uber and its idea of matching “I want a car” with “I have a car”. How does a more structured Taxi Network compete with that?  The answer so far has been scare tactics, calling in political favors, strikes,  and seeking to burden the other the same way they are regulated. (And conveniently ignoring their government-sanctioned monopoly protection in many cities.  It hasn’t been in lower fares, more drivers, cleaner cars, broader service.)

We see this dynamic in politics, society, and business. At the Macro Level of commerce, Nobel Laureates Oliver Williamson and Ronald Coase outlined how transaction costs impact the design of firms and small teams. The focus of their innovative insight was the to address the gap between theory of the “pricing instrument” as the central coordinating instrument to a market (macro) and the role of specialist/entrepreneur in defining how companies are organized. Employees who are “Smart creatives” are attracted to where they can deliver value and interact within a supporting community (not in driving a commodity). The Fortune 500 have never been more challenged to justify their current structures in a world where speed and talent matter more than their brands and the barriers to entry they’ve been able to create over time.

More detail in blog Internet 3.0: Collaboration and Communities

Hypothesis #5 – New Clusters form on “intersections of value” (ex Uber) with primary margin capture by the orchestrator of the interaction (ie the center of the star network).

No network can start without at least 3 nodes (participants). Two people are friends; three people are a network.  Each node has limited energy to connect with others, so networks often form around  specific value propositions (see Building Networks and Openness). Existing networks are rigid, as participation was around a specific value proposition, and people and companies can build new ones as new specific value propositions arise. Some of these new networks may have some of the same participants, but it becomes a juggling game. Google’s real data advantage is that it allows an infinite number of  “open intersections” between heterogeneous nodes (consumer and everyone else). Commercially, this allows Google to use data to make its products better, make consumers happier,  and thus gain more use—the “virtuous circle” which provides for further Data Economies of Scale (and profits!).

There are thousands of companies that have BETTER, deeper, more targeted data than Google, but that data remains locked up because data use and pricing can’t be controlled. These constraints ARE NOT TECHNICAL. Existing Companies (data owners) use this data to enhance their own CORE value. New companies seek to rewire connections to create new value.

Just as with Google above, data enables differentiated products and experiences. Google touches the average US consumer more than 20 times a day and must deliver value in each of these interactions. Google’s “open” advantage is search, as search signifies “intent”. Looking for hiking boots? Must be headed on a trip. Searching on flu? Someone in your family is sick. Facebook’s “open” advantage is social graph. The social graph captures behavior and preferences for many networks (see blog).

How does data relate to other data? For my European friends seeking to hobble Google,  help me here!  Vodafone and Telefonica have far more consumer behavior data. MNOs know consumer search word, location, apps they have installed, who they call, where they live, where they bank. This MNO data set on individual consumers is of far greater value than what Google has. The problem is that MNOs have not created services that USE the data to improve their networks other than where to build towers. So Google’s real EU crime is using the data that your EU companies already have, and creating free services that your customers love. If the EU wanted to enable competition perhaps they should force their own MNOs and ISPs to open up the data pipes to destroy Google’s competitive advantage. Of course this would be a nightmare as there is no portability of trust across domains. No mechanics for the definition and exchange of relative value. Thus Google is guilty of an additional sin… having consumer trust!

For intersections to occur there MUST BE TRUST, CONTROL, and Market forces. For anyone to compete against Google they must find one or more partners (see Clusters and Collaboration).  My chief Scientist Dr Rod Cook hypothesis is that most of our interactions are with small networks. The challenges for [small] data owners are trust, domain, discovery, and relative value .

In last 10 years, most enterprises have gained a new sense of awareness of their own data capabilities—and their need to UP THEIR GAME. Their data, while tremendously valuable, is not unique enough to compete with Google’s on many fronts. They have also realized that their ability to become a channel (think Card Linked offers, or MNO led Ad platforms) is limited because of their reach, value and trust outside of their core. In essence EVERY COMPANY is a “small data” business next to Google. And many of them seem happy just to sell their data to Google and others.  They cash the one check rather than learn how to cash many checks, and more importantly, how to write checks to get the data they need to improve their own offerings, and build their own better reputations, and trust.

Unfortunately, the industry that can immediately act on multiple data sources (Digital Advertising) is most focused on enhancing the capability of its own core (not that of the data owner). Ad people tend to view data as an asset–something to use and monetize. Data Owners view data like their children.–willing to let them play if it adds value to the overall family objection,  but only for a short time (eg, the kids can run a corner lemonade stand on a Saturday afternoon, but not take a job at the local sewing factory).

Incremental data will provide incremental value, but there are a lot of issues.  Who get the incremental value? Who can use the insights gained? Who bears responsibility when bad things happen?   While Google may know what you did online (10% of commerce), they have only a marginal idea of what you do in the physical world.  Facebook gets a good idea of what you want to tell others you did in the real world.  MNO’s know where in the real world you were at a certain time.  Retailers, mobile operators, banks, all have tremendous insight and data in the physical world, but are constrained in using it collaboratively (see Mobile Economy). The challenge for the “small data” that each of these companies has, is determining where, when, with whom, and why you share it.  One big objective of sharing is when it’s not for just monetization of the data—getting the one-time big check– but supportive value creation for your company and those in all of your many networks.

This challenge is further exacerbated by multiple “languages” for identifying a customer against millions of partners with across complex ownership, customer and permissions. Five companies know the same consumer by a different email address or phone number.  Now you know why most digital marketing is so bad—people have to guess that is the same as is the same as (123)456-7890. The first party data is locked up and proxy data rules the day. So how do we break the log jam? (Commerce Signals)

#6 Networks of heterogeneous participants (Google) have greater value potential than that of homogeneous participants and uses (Mastercard, eBay, Amtrak). Logically if value is unlocked at the intersection of participants, the intersections of dissimilar entities have greater value than that of peers.

A participant in a network has a purpose/use. The greater the number of uses, the greater the number of transactions within the nodes AND the greater the number of intersections on seemingly uncorrelated data (ex consumer intent models).

#10: Blockchain

Nothing in finance, anonymity, or in government taxation is as disruptive as blockchain! How distruptive? How many intermediaries are there in a typical stock market transaction? Must read brief by Tim Swanson on topic and his picture below illustrating the fundamental network changes that occur when trust evolves outside of a market onto a specific instrument (ie self-authenticating transaction). I don’t need to trust the transaction counterparty if I am assured he owns the asset and is transferring it to me. Blockchain is a great example of how technology can change a hub and spoke “star network” into a scale-free network.


A distributed ledger still requires “a Market” for pricing. However brokerage, order execution, clearing and settlement and reporting all go away. An open “scale free” anonymous network causes nightmarish problems for regulatory reporting and off-market trading.   Blockchain is transformative in that it solves a trust, identity, distribution and settlement problem. Imagine applying the following paradigm to commerce “I don’t care who you are if you have defined Product/Commodity X that you can Deliver to Y”.

So much of our financial infrastructure has been built on trust, risk management and reputation of exisiting frameworks and players. Can technology really replace it? My answer is yes, IF CURRENT PARTICIPANTS SUPPORTED IT. And that consumers and users trusted those who say they trust it.  The control points for the traditional markets have significant “teeth” however, particularly stock ownership records and reporting requirements for current participants. As one regulator told me, the “problem that blockchain solves is disintermediation, and that isn’t a problem that current stakeholders need solved”.  I will have to address more on blockchain at another time.

Parting Thoughts

How long do partnerships take? How many of them turned out the way you imagined? (How many of you have been divorced?) Its very hard to innovate when you aren’t in control of your delivery channel, particularly when your product is a commodity and the value rests with the network  (think ChasePay). Margin and Trust is quickly flowing from the nodes (Banks and Retailers) to Orchestrators (Google, Facebook, Amazon , …). The only way to compete against this dynamic is to collaborate with non competitors (ex Amex’s Plenti). Value is at the intersections and everyone must work to create them during the rewiring of commerce. This is not a technology problem, but a business one.

I’m thrilled to have a fantastic team of employees and partners focused on this.


1 thought on “Transformation of Commercial Networks: Unlocking $2T in Value”

Leave a Reply

Your email address will not be published. Required fields are marked *