Palantir is one of the most fascinating companies in the Valley because the core purpose of the company is data-analysis based intelligence products and services — a setup so close to the nature of operations research. Palantir occupies a weird service space between Tool-kit Developers like Matlab and Mathematica, Consulting firms like ZS and Booz, and IT service companies like Wipro.
Evidently, Palantir does a lot of data-analysis for the government agencies reportedly on terrorism, and hence is under constant scrutiny by journalists and civil liberties organizations, and it is famously opaque. So, it is always interesting when some light is shed on its business. The article in WSJ is informative in some ways. The authors are skeptical on Palantir: The article is titled, “Palantir Has a $20 Billion Valuation and a Bigger Problem: It Keeps Losing Money” (subscription required). The article is also polemic, but the statements by CEO Karp, — “a self-described socialist”(!), according to the article — do not help.
“This is not a science,” he says. “We are a colony of artists. You cannot go to Basquiat or Monet and say, ‘Well, that painting didn’t capture the time.’ ”
For sure, there is a lot of science to statistics and data science behind all the effort.
My understanding from the article is the following:
- Perks have gone down. Cost-cutting and Belt-tightening is occurring. This is OK, and not newsworthy.
- There is significant pressure to go public, but there are natural worries about business model post-IPO.
- Issues in navigating or pivoting from government to private businesses.
- A realization is need for good sales team.
- There is a perception of stiffening competition (the “Uruk-hai”).
Two observations going forward on Palantir. (Thanks to the two individuals that I talked to with previous experience at Palantir, that helped me understand the intricacies).
1. Product-Service Bundle
It appears to me that Palantir is realizing the value of tailored services.
The biggest differentiation advantage for Palantir team is their products: Gotham which integrates and secures data across enterprise, and Foundry, which analyzes the integrated data for intelligence and patterns. Products with flexibility to plug in are neat, and they differentiate Palantir from highly service oriented consulting firms.
(See how Foundry can be used in Supply Chains for Chrysler. Source: WSJ & Palantir).
However, no product is purely a product. As one can easily think in the case of Automative supply chains, products sold to enterprises don’t “plug and play” despite best efforts. In the Chrysler example, there are many legacy issues related to how dealerships adopted various practices.
(An Aside: That’s a PT Cruiser in the picture! A very hot product in the aughts, not anymore. That kinda of dwindling sales information can be identified by Foundry, but what is the best intervention for Chrysler?)
We don’t know the detailed tool-kit level intricacies that make shifting from military and law enforcement deployment harder. We don’t know much about why JP Morgan and Hershey have canceled their contracts. However, I would speculate that many firms are not just satisfied with product solutions. In addition to products that they deploy, firms need high touch services.
Palantir has had a traditional dispreference towards the Sales Force. Purely engineering genius teams are great for startups and research labs, but a successful organization has to understand the usage issues for customers.
In this respect, I think Palantir would be best prepared to build a Service Arm (by deploying/acquiring a boutique consulting firm), and then use them to create a seamless product-service bundle.
2. Emerging Competition
Any person who considers themselves a Tolkien fan, knows that palantiri are not unique. In fact there are seven of them, owned and employed by many including Aragorn, Denethor and Saruman. Internet jokes have speculated on whether Palantir as a company resembles Denethor or Saruman, which is besides the point.
Crystal Balling, simulating and data visualizing are compelling but not unique skill sets, especially in the talented engineering workforce in Silicon Valley. Palantir has had the leg-up because of early start, high profile, aggressive successive investments and cash advantage, but as they make efforts to go public, it is only reasonable to wonder about competition in this space.
The competition is not Uruk-hai, an army of half-orcs with primitive weapons that fights to the bitter end. Competition will be from other firms with the same skill set and possessing the data-crystal ball — the legends have there are at least four such “seeing orbs” within Gondor, the land of men — so I do expect this space to get more competitive. I look east to China where a ton of entrepreneurial activity in the big data space.
End Note
One of the fascinating tidbits in the article is the graphic (below) on the most valued “unicorns” in Silicon Valley. Valuations are generally optimistic for firms, and particularly so for firms like WeWork and Pinterest in my view. Yet, the data is astounding.
And yet, if one company that does belong here on the list, in my view, it is easily Juul. Juul is a reflection that a few have never stopped prospecting for easy gold out west. I will write about Juul in a later post. I also cannot meaningfully explain the extraordinary gap in valuations between Uber and Lyft. We will see how that race plays out.
If you like what you read, you can subscribe to the OWL Blog for occasional newsletters. No spam. I respect your privacy.