Feature

Smith + Crown’s Industry and Sector Classification

There have been many efforts to classify startups and blockchain projects by industry. Outlier Venture’s token ecosystem classification is particularly insightful. Classification serves multiple purposes, including helping cluster projects working on similar problems, identifying peer companies, and tracking growth over time. This helps identify which sectors are particularly dynamic, indicating a shared intuition of a particularly lucrative blockchain application.

Industry classification is not as straightforward as might initially appear. The most comprehensive efforts to classify business activity are usually undertaken by governments. In the United States, industries must register according to the NAICS code system, which generally resembles the International Industry Classification system.

Classification: Means of Production v. Product?

There is one key decision in such classifications that the lay observer usually misses. Most systems categorize companies according to their means of production rather than their product. An example would be how Uber is classified as an information company rather than a transportation one, or Airbnb as an information company rather than an accommodation business.

From the standpoint of governments, this makes sense. Means of production is a better indicator of employment needs and industrial input–two areas over which governments have immense influence. Manufacturing cars requires different types of workers and resources than designing software to support car manufacturers. It has the added (and unintended) benefit of remaining mostly stable on a per-company basis: the software company could begin selling software to distributors (a different sector).

This approach also has several drawbacks: it is very poor at making sense of the technology sector, particularly information technology and its impacts throughout the economy. Under current production-oriented classification (like the NAICS system), a software company solely supporting car manufacturers would be considered a software company. If a car manufacturer acquires that company to perform the same services in-house, it would seem like the manufacturing sector grew and the software sector shrunk.

In addition, the actual products are important to keep track of as a means of identifying market changes. Governments have the luxury of tracking this through product sales numbers, but this isn’t always feasible. It also makes it difficult for companies to identify competitors or strategic threats. Airbnb undoubtedly has an impact on the accommodations industry. Finally, the commitment of means of production is also likely short-lived. On a certain level, every company is an information company.

Classify by Means of Production Classify by Product
  • Likely better reflects labor force changes
  • Better reflects industrial inputs (land, energy, materials)
  • Classification can follow companies across multiple markets
  • At a certain point, seems like a false distinction
  • Difficult to answer business strategy questions, such as identifying competitors or threats
  • Cannot make sense of widescale industry transformation

Smith + Crown’s Classification

Smith + Crown has considered this issue carefully as we develop our own estimations of blockchain-related activity across a range of industries and sectors, and we have implemented to following approach to classification.

  • Classify according to target market. What industry is the company trying to disrupt? In other words, if it is a payment system for marijuana, it operates in the drugs and alcohol sector.
  • If the project’s market is industry agnostic, classify according to its product. For example, if it is a general purpose payment system, it operates in payment processing sector.
  • Classify according to several new sectors that are specific to the blockchain industry, including smart contract platforms and prediction markets.
  • Finally, use the NAICS code industry system as a backend and maintain a cross-walk to new industry and sector classifications.

We have done so for the following reasons.

  1. Classification by means of production would be very difficult and usually arbitrary. Ultimately, most companies are applying blockchain technology in one of several ways: an immutable and auditable ledger, a digital currency, autonomous and auditable software, and an economic system for crowd contributions. Even these distinctions are rough and not comprehensive.
  2. It is a more coherent story of disruption. Classification by (arbitrarily chosen) means of production wouldn’t shed much insight into how blockchain technology is actually being used to disrupt traditional industries.
  3. It helps identify peers and competitors. This is important for assessing the relative merits of different tokens.
  4. It is more straightforward. Today, most blockchain companies are small. While they will pivot or expand their services, in their early stages they have a distinct focus that is usually easy to identify.

General Trends

We have developed a classification system for the various sectors into which blockchain-based companies are focusing their energies and integrated those results into our existing project database. This allows a number of novel insights into the evolution and current market conditions of the crypto industry, and can be particularly fruitful when evaluating ICOs. A few examples of the kinds of analysis this system allows us to develop have appeared in some of our recent  research reports, for example our Covesting, Solar Bankers, and NapoleonX reports, and we have much more to share in the coming months.

For now, an initial look suggests some of the ways information about sectors can shed new insights and improve our understanding of the larger space. The chart below shows our sectoral classifications, with the size and shading of the respective boxes representing the amounts raised by each sector. 

We will be following this up with an explanation of our sectors and hope to solicit feedback from the community.

1 Comment

December 28, 2017 at 1:39 pm, James said:

I am glad to see that someone is making an attempting to come up with classifications that work. Thank you.

Reply

Leave a Reply

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