A.I./Blockchain Adoption and Technology Substitution

Tue, 02 Apr 2019 00:00:00 +0000

Artificial intelligence (A.I.) and blockchain are the hottest new technologies right now. It would be fun to think about the future of them (see the patterns, draw the inference, and make predictions). Would A.I. and blockchain be the next “internet,” which has disrupted many traditional business models and creates many new giants across the globe? Or, since businesses have learned a lesson from the internet disruption, A.I. and blockchain will be merely incorporated into the existing technology terrain as an expansion of opportunities by the incumbents? Do A.I. and blockchain have the same fate?

I’ve been leaving an eye on the adoption/acquisition of new technology in the banking sector (since one of my dissertation chapters was on banking acquisitions). Here is some latest news.

The banking sector has a bright prospect incorporating the A.I. in its six core functions.

JP Morgan has adopted cryptocurrency in the payment businesses.

Wells Fargo and Mastercard CEOs express doubt about blockchain.

PayPal made a major investment on the blockchain.

Could there be a unified explanation for what is happening right now? Well, one may associate the topic of technology development with Clayton Christensen’s ‘innovator’s dilemma’ assertion, which predicts emerging new technology will quickly take over declining old ones (often depicted as the double-S curves). It’s the good management–trying best to hear about customer needs, play with competition carefully, and align resource allocation with calculated risk/benefit profiles–who prevent the incumbents from adopting new technologies. Then shouldn’t managers try their best? According to Clayton Christensen, it is difficult for good management not to do good–it is difficult for the incumbents to utilize new technologies even they tend to spot them earlier than startups. At the initiation stage of new technology, incumbents can’t risk their existing market share and customer relationships with unproven models, whereas startups have little to lose. The prediction is: startups disrupt incumbents.

However, things are different right now. The boundary between new tech. and old tech. are melting down; they are intertwined ecosystems rather than isolated communities. Incumbents right now have taken deliberate effort and allocate resource for emerging technology. They are agiler and more determined to disrupt themselves before being disrupted by others. If important pre-existing conditions are different, it might be necessary to modify the predictions from Clayton Christensen’s double-S curves .

First, the compatibility between incumbents’ existing technology and emerging technology should lower the likelihood and rate of disruption. Let’s consider two cases here. One is about different incumbent technologies: as compared to traditional banking, the higher level of digitization of PayPal allows a higher compatibility potential with blockchain (consider internet and blockchain don’t rely on a central authority/agency as traditional banking does). The other is about different emerging technologies: as compared to the blockchain, A.I. allows a higher level of compatibility with traditional banking (both banking and A.I. are aimed at decision/prediction accuracy and efficiency.)

Second, the intensity of the incumbent technology expansion should also lower the likelihood of disruption. This factor is not only related to incumbents’ existing resource base but also has something to do with management. A recent paper in SMJ ¹ has updated the “double-S” framework by incorporating the “incumbent technology expansion” in predicting the rate of substitution and found empirical support for the model’s prediction efficacy.

Conclusion: since A.I. tends to have a higher level of compatibility with, and also face a stronger expansion threat from, the incumbent technology than blockchain, it’s predicted to have a lower likelihood/rate of disruption. However, contingencies should also exist due to the significant variance in incumbents’ technology, resource, and management conditions.

Adner, R., & Kapoor, R. (2016). Innovation ecosystems and the pace of substitution: Re‐examining technology S‐curves. Strategic management journal, 37(4), 625-648.↩

System Mapping in Business Modeling

Thu, 28 Feb 2019 00:00:00 +0000

In the past several years, I spent quite some time coaching college students in the business major on their individual/team projects. These include business model design (BUS146), strategic audit (BUS109), and marketing plan (BUS103). In each quarter, there are around 100 students enrolled in BUS103/BUS109, and 50 in BUS146. And they typically select different companies and different entrepreneurship ideas to work on.

At some point, I realized it is crucial for them to quickly “model” a business so they can have a very clear picture of how businesses work. I also usually suggest they put all their thoughts on one paper. So they can have the big picture without missing out any important information. This is where “system mapping” can come in handy.

It involves three simple steps. First, brainstorm the essential components in a business system. Second, identify the causal linkages and loops. Third, mark the positive/negative relationships (and the leverage in the system). I find that most students can master this skill very quickly and can generate a good graph with some practice.

I also push for simplicity, as I notice the more I ask them to succinctly define the components, the more they move away from perceptual experiences to critical thinking. Such a map can be a helpful starting point for us to keep questioning ourselves on further facts, proofs, and underlying mechanisms.

Here is an example of a generic “precision online marketing” model (e.g., Google, Facebook, Youtube, etc.), made with iGraph.

business model on Ray Yang, Ph.D.

A.I./Blockchain Adoption and Technology Substitution

System Mapping in Business Modeling