Interesting! I did not hear about Philyra before today, but I’ve spent the last hour trying to understand how AI can be used to create materials with new properties (fragrance, in this case).
It looks like Symrise and IBM use ‘distance’ between raw materials to create the most novel fragrances. Does novelty always translate to appeal in fragrances? I wonder how they can further couple this process with human perfumers to create a feedback loop that allows them to manufacture new combinations of fragrances without having to resort to distance as a measure.
This is a radical idea , but can Philyra integrate ‘fragrance sensors’ (like electronic noses – https://en.wikipedia.org/wiki/Electronic_nose) into the learning process so that this system can constantly generate new perfumes with only minimal human intervention from the perfumers?
With IoT devices constantly coming under attack by hackers, I think it is understandable that parents are skeptical about exposing their children to an IoT device with recording facilities. However, with better IoT security solutions being invented every day, Hello Barbie (and other similar devices) may well be accepted by parents in the near future.
I also wonder if the branding of the doll may have contributed to its failure. As parents are increasingly keen on introducing their children to educational toys early on, could ‘Hello Barbie’ have instead be designed and branded as an educational doll instead? While child therapy is indeed a great use case, the market segment may be significantly smaller than that of educational toys.
Another risk that should also be considered is the likelihood of bias. Since ‘Hello Barbie’ is a trailblazer in speech processing for children, some of the bias concerns around AI are likely not completely addressed and understood. Thus, the exposure to negative PR and regulatory threat is significant.
Great article! It is exciting to see robots change the way stores and warehouses manage inventory. I am curious why Bossa Nova chose to specialize in retail store inventory management, and not warehouse management, which is likely a bigger market.
Also, how scalable is Bossa Nova’s solution? Since its robots rely on LIDAR maps for navigation and localization, how easy or hard is it for the robots to adapt to a new retail setting and to understand new product SKUs? Are the LIDAR maps also generated and refined by Bossa Nova’s robots, or does it incur additional costs?
Amazing post, Sarkis! I was intrigued by Salesforce’s acquisition of Tableau, and your in-depth analysis elucidates the rationale behind this deal.
While Salesforce is benefiting hugely from Tableau’s massive data processing, cleaning and visualization capabilities, how does Tableau use data to better its services and products? For instance, since they have a large customer volume, can they leverage their customer data to improve their dashboard and visualization capabilities? Also, where does data ownership of the other (non-Salesforce) customers lie, and how has that changed with the deal?
Since the success of this deal may inspire other similar vertical integrations by big tech (or other industries) players with their data capabilities suppliers, I am very curious to understand the regulatory and legal repercussions of owning and/or using customer data before the integration?
Excellent post, Keagan!
In addition to art democratization, I believe that digitization of art may have other desirable consequences –
– Data-driven curation. As Jackson mentioned above, museums expend many valuable resources over the curation of their collection, which is a constantly ongoing process and is driven almost purely by the curator’s intuition and experience. Using data to automatically curate and funnel relevant art from their expansive archives to users’ eyeballs will help the museums generate greater value for the user.
– Pervasion of art into social media. The stock images market is valued at several billion dollars today and is rapidly growing. Democratizing art will likely allow the stock images market to more fully consume and build a synergistic relationship with art, thereby benefiting both aficionados of art and users of stock images.
– Motivating a younger global generation of artists. When art is on social media, more millennials can experience, and be empowered and motivated by art. The global reach of social media can also be extremely powerful in drawing more prospective artists from around the world.
Such an incredible idea! It makes me wonder how many other applications can benefit from an amalgam of data from several sources. In particular, many underexploited niche markets with low TAM can likely benefit from finding and aggregating complementary data sources.
If Rachio relies so heavily on data from external sources (with potentially constantly changing APIs, or even no well-defined APIs), how do they ensure sustainability of this strategy? Will they have to rely on teams of all-star engineers and data scientists to constantly recalibrate changing datasets and ensure that the results seen by all the customers are repeatable and consistent?
Thank you for this excellent post! As someone who grew up in India, I can resonate with the difficulty in finding a reliable hotel every time I traveled, and the OTAs did not help. OYO is indeed an excellent idea in a market that is very price-sensitive, but still desires standardized hospitality experiences.
OYO has pivoted its business model from being more of an aggregator, where it sources, leases and then operates rooms, to a franchisor (https://www.feedough.com/business-model-oyo-rooms/). How, in your opinion, do the cross-side network effects change with this different business model? My hunch is that they should be diluted, because franchising needs higher capital expenditure and requires more stringent compliance and control systems, and this, therefore, reduces the incremental value derived by a hotelier when more users join the platform. If so, how will this affect OYO’s growth strategy?
The network effects will likely further be weakened by the network structure. Since the hotels are likely to be concentrated around urban centers, will the local isolated clusters potentially undermine the robustness of this network (https://hbr.org/2019/01/why-some-platforms-thrive-and-others-dont)?
Also, I’d be interested in knowing how much of a problem disintermediation can be, particularly for regular users with predictable travel needs and plans. Can OYO think of ways to reduce this disintermediation?
It is very exciting to see Coinbase at the helm of the cryptocurrency revolution, and as you describe in your post, they seem to be using several platforms to do so. In particular, it is interesting to see how they are betting on only a handful of cryptocurrencies, as opposed to competitors like Binance and Kraken.
I find your analysis of the multihoming threat to Coinbase very interesting! While I agree with your suggestion that Coinbase is invested in the growth of the overall cryptocurrency market, I do think multihoming can still be a credible threat. The number of APIs and platforms in the cryptocurrency space is rapidly proliferating (https://blog.nomics.com/essays/cryptocurrency-market-data-apis/), and unless the system can agree on a single standardized API, interconnectivity and interoperability may be stunted, and that does not bode well for the crypto space. In my opinion, as one of the market leaders (depending on how we look at it), Coinbase should actively reduce multihoming by collaborating with its competitors (through consortia, for instance) or by leveraging its network effects to shove them out of the market.
Also, it is interesting to see how same-side network effects are just as important as cross-side network effects in several of the Coinbase platforms. By enabling peer-to-peer transactions, Coinbase seems to have mastered the mantra for rapid scaling through network effects.
Love this writeup!
As a Kaggle user myself, I wonder how Kaggle deals with multihoming. Since several other companies are competing in the online collaborative coding space (like SPOJ, Topcoder), and many more are coming up with interesting differentiating factors (like Numerai), how can Kaggle prevent its user base from multihoming?
Also, I’d be curious to understand the network effects in play better. I can imagine there must be strong cross-side network effects between users and companies, but the same-side effects seem to be weak at best. What is their scaling strategy, in the absence of organic scaling drivers like network effects?
Great post, Riddhi! TripAdvisor is a savior, and I almost cannot imagine traveling without the friendly tips and reviews on TripAdvisor.
In my opinion, TripAdvisor has to strike a very careful balance between reducing barriers to entry for new users and businesses, and increasing barriers to prevent proliferation of fake profiles which can obliterate brand equity and trust. How do they identify the point of equilibrium? Also, presumably, they would want to err on the side of higher barriers to entry, because the brand equity they have built over several years is more important than a marginal increase in user volume.
Another question this write-up leads me to ponder over is the standardization of ratings across geographies. How can TripAdvisor ensure that a 4-star rating means the same in Bali, Indonesia as in Thule, Greenland, so that customers find the ratings reliable and consistent?
Excellent post! I am very excited to see how Quantopian evolves, particularly given the recent change in investment leadership and product strategy.
A couple of quick questions – how do you think they can reconcile the community-generated alphas with their internal investment strategy? Since community alphas are very broad in risk and industry exposure and return profile, should they have an internal investment strategy and filter the alphas they get to a subset that aligns with their thesis? Alternatively, can they let the community jointly and dynamically determine the thesis, and focus purely on synthesizing the raw alphas to maximize risk-adjusted returns?
Also, is there an equivalent of slant (https://econbrowser.com/archives/2010/02/what_drives_med) and bias on this crowdsourcing platform? For instance, since early adopters of technology are likely over-indexed on the platform, the alphas might reflect an inordinate bias towards the technology industry, perhaps?
Thank you for the great post! Having been on the founding team of a fashion startup that marketed primarily through Instagram, I find Fashion Nova’s marketing strategy very impressive, and I am glad I got to read this enlightening piece on their digital marketing tactics.
One question I have is – how defensible is their digital marketing strategy? Since Instagram marketing is a very saturated and competitive space, customer switching costs are low, multihoming for customers is incredibly easy (that is, they follow multiple brands and multiple influencers), and influencer popularity and relationships with brands are often short-lived, what prevents new niche entrants with a similar price point from competing with Fashion Nova? Also, since their business model is reliant on a large inventory, how scalable is their distribution?