Monday, October 4, 2010

Review: Suborbital Market Overview and Application of Disruption Theory

In a recent paper, Ken Davidian of the FAA Office of Commercial Space Transportation, and Jeff Foust of the Futron Corporation have applied Clayton Christensen’s Disruptive Innovation Theory to the suborbital launch industry, predicting the impact of RLV’s on the suborbital market by describing the impacts from multiple technology introduction strategies. The resulting paper provides significant insight.

First a quick summary of Christensen’s Disruptive Innovation Theory (think of this as three strategic options for RLV companies entering the suborbital launch market):
  1. Sustaining Innovation: As an RLV company, enhance one of the current sounding rocket capabilities. Fly higher, reduce g-forces on payloads, reduce cost of launch, reduce purchase-to-launch cycle times, etc. Competition from incumbents will be high. Marketing Risk will be low (you already know the market exists).
  2. Low-End Disruptive Innovation: As an RLV company, offer a lower price than sounding rockets and offer an inferior product (e.g. by not flying as high as a sounding rocket - early RLV’s will offer fewer minutes of quality micro-gravity). Competition from incumbents will be low since this strategy steals the lower margin portion of the market (those customers wanting “a deal”). The incumbent will instead focus on the high-margin portion of the market. Marketing Risk will be low.
  3. New Market Disruptive Innovation: Offer a new capability not offered by sounding rockets. Fly people, return experiments at mission end, fly more than once per day, etc. Incumbents will not be able to compete in the near-term in most cases since current sounding rockets do not offer such capabilities. Marketing risk will be high since new market disruptive innovation must pursue “non-customers” – those not currently served by sounding rockets.
Here are some nuggets from the paper:
  • The paper argues in favor of Low-End Disruptive Innovation as a preferred strategy for Government customers to support RLV operators – encouraging use of RLV services even before the capabilities of such RLV services fully meet Government needs (or fully matches sounding rocket capabilities). The authors argue this is the best way to help grow a sustainable industry.
  • Quoting studies from Christensen’s book, Innovator’s Dilemma, new RLV companies would garner a significant first mover advantage by pursuing either of the disruptive innovation strategies mentioned above: new entrants in an established market were successful only 6% of the time while “first mover” new entrants pursuing disruptive innovation strategies were successful 37% of the time. The first move advantage is large!
  • The paper considered “low-end” suborbital markets to be: earth remote sensing, astronomical & atmospheric observations, technology demonstrations, educational payloads, and novelty payloads that can be performed with only one minute of quality microgravity.
  • Since 1942, suborbital sounding rocket altitudes are grouped into three categories: 100 kilometers (4 min of microgravity), 300 kilometers (10 min of microgravity), and 300-1500 kilometers (astronomical observation mostly) – with the majority huddled into the 100-300 kilometer range.
  • The paper predicts according to Christensen’s Disruptive Innovation Theory, early RLV’s will use proprietary technology and be highly integrated, but as more RLV competitors join the market, RLV products will become more modular.

If I had a critique, it would be:
The authors assume the suborbital market would have a large enough “high-margin” market segment to allow incumbents to thrive even while surrendering the low-margin segments to RLV’s. Let’s assume the high-end segment of the suborbital market is any mission significantly over 100 kilometers and the low-end segment is 0-100 kilometers. Looking at the powerful graph on page 11 of their paper, it is clear that the suborbital market is already disproportionately skewed toward the “low-end” portion of the market (although the higher altitude market does appear to be growing). Just by eyeballing the graph I would estimate 35-45% of the suborbital market is 100 kilometers or lower. Would incumbent sounding rockets be able to charge a large enough premium for launches above 100 kilometers to justify losing 35-45% of the market and not retaliate through lower prices?

Remember, I am not doubting the success of RLV introduction into the suborbital market. Instead I am raising doubts on one of the authors’ key conclusions that the sounding rocket incumbents will flee up market rather than retaliate with lower prices. I am not sure the market is large enough for the incumbent to do that. If not, I would expect sounding rocket companies to lower prices to compete with RLV’s even up to 100 kilometers. If, as an RLV operator, you agree with my critique, “New Market” disruptive innovation strategies (although higher marketing risk) may actually make more sense since sounding rockets would not be able to emulate the new RLV-enabled capabilities (in the near-term).

I am a big fan of Clayton Christensen and believe his disruptive innovation theories (especially low-end disruptive innovation) would more perfectly apply to an analysis of Nanosat launchers as a disruptive orbital launch technology. But with that said, Davidian and Foust’s paper provides a great overview of Christensen’s theories and provides significant insight into the future of the suborbital market. And let’s not forget, this is primarily a government paper written to provide recommendations to the US Government on how best they can promote this industry – I do very much like that!

13 comments:

  1. Colin,
    I don't think the incumbents can compete on price without completely restructuring around how a new entrant would operate and Clayton has shown that incumbents almost always fail at that. But price isn't the only area where incumbents can't compete. The customers in that tight 100km knot also want fast turn around time and low integration costs. The incumbent's systems have extremely high g-loads that require significantly high engineering and integration costs.

    As Clayton says, it is easier for the new player to go up market and eat the incumbents lunch than it is for the incumbent to go down market.

    ReplyDelete
  2. Michael:

    You say, “price isn't the only area where incumbents can't compete. The customers in that tight 100km knot also want fast turn around time and low integration costs. The incumbent's systems have extremely high g-loads that require significantly high engineering and integration costs.”

    I agree. Fast turn around times, low g-loads and low integration costs would be excellent examples (of many) of New Market Disruptive Innovation. Ken and Jeff’s paper argued in favor of Low-Market Disruptive Innovation where RLV’s could offer a low cost service at the bottom of the market and steal customers who were over-served by the current sounding rocket service. This disruption strategy is meant to describe a way for new entrants to avoid price competition with incumbents.

    The problem is the suborbital market is too small for sounding rockets to ignore RLVs (Colin’s big assumption). There is not enough high-margin up market for them to focus on. So what happens? I believe sounding rockets will fight for those 100km customers. In the short-term, they can’t compete by lowering g-loads or integration costs, so how can they compete – perhaps the only tool available to them will be price. They may not be able to lower their price very far, I honestly don’t know the profit structure of these firms.

    But the paper argues in favor of Low-Market Disruptive Innovation as the best strategy for the suborbital RLV industry. Since that strategy will not avoid the price competition seen under Sustaining Innovation, I believe focusing on New Market Disruptive Innovation will do more to benefit the RLV companies even with its corresponding higher Marketing risk.

    Fast turn around times, lower g-loads, lower integration costs – yes, this is the way to go!

    -Colin

    ReplyDelete
  3. Colin,

    Thanks for the review and the thoughtful comments. This study, and others like it*, are intended to be iterative in nature... used as a starting point for more in-depth and deeper discussions on the topic. Reviews like yours (and Mike's... hi Mike!) will only make the next version of this paper that much better.

    Thanks again!

    Ken

    *I'll create a page on the Commercial Space Wiki to collect this study and other studies like this where enthusiasts like us can critique each others' work.

    ReplyDelete
  4. Colin,

    Another paper you might be interested in:

    http://commercialspace.pbworks.com/f/IAC-10.E6.3.1+ISA+of+Commercial+Crew+to+Orbit+FINAL.pdf

    This paper by Brad Cheetham takes an first-look at the emerging Human Transport Vessel market through the lens of Porter's Industry Structural Analysis.

    Ken

    ReplyDelete
  5. Ken:

    I will definitely review the article via the link you provided.

    I can’t thank you enough for your paper and your work at the FAA. Keep it up.


    Colin

    ReplyDelete
  6. Ken:

    I wonder if Christensen or others have performed analysis on market size, i.e. at what market size do the rules of low-end market disruption no longer apply?

    Such analysis could address the validity (or lack of validity) of my critique – namely, is the suborbital market large enough for sounding rocket companies to flee up market or would the small market entice the incumbents to behave as they would under sustaining innovation?



    Colin

    ReplyDelete
  7. I will read these papers with great interest and may end up citing them in the Astronauts4Hire business plan.

    ReplyDelete
  8. Colin,

    I've been thinking about all the comments here and for the *next* iteration of this paper, or for my next DT paper on another market (I assembled most of the data for Earth to LEO Cargo Launch Vehicles about a year ago, so that might be next), I think I need to state that this analysis pretty much assumes perfect competition. A follow-up to this paper could be a description of all the ways that the competition *isn't* perfect, including many of the points mentioned above.

    Ken "Still Thinkin'" Davidian

    ReplyDelete
  9. I have had excellent discussions this week about the suborbital RLV market. The thoughts below are thanks to many of you who have been dialoging with me off-blog.

    What-if sounding rocket suppliers cannot lower price? Does this change my view on the validity of low-end disruption as the preferred disruption strategy for the RLV suborbital industry (Ken and Jeff’s assessment)?

    If a NASA sounding rocket costs $5-10M per flight and if you assume there are eight 10-kg payloads on board each flight, each 10-kg payload is paying $600-1200K per flight. RLV’s are talking about charging $50K per 10-kg (maybe less). The difference in price per payload is staggering: $1M to $.05M. Even if we assume 10%, 20%, even 30% margins for sounding rocket suppliers, they simply can’t match the RLV price (Colin’s second big assumption in this thread). So under these conditions, sounding rocket providers may have no choice but to focus on high-margin customers. I still think the market is too small for sounding rocket providers to relinquish the lower-end market share willingly, but if the RLV price point is so low that sounding rockets could not retaliate by price-cutting and stay in business, then the low-end disruptive innovation strategy advocated by Ken and Jeff may make more sense then I originally thought.

    However (and this is a big “however”), asking the following question may help determine the best disruptive innovation strategy, “Why keep RLV prices so low - $50K (or less)?” For low-end Market disruption you keep prices low to avoid incumbents responding with lower prices of their own (price war). But in the paragraph above we agreed that sounding rockets most likely cannot lower their price (substantially). If the incumbent cannot lower its price of $600K for a 10-kg payload, why shouldn’t RLV operators set their price at $100K, $200K, or even $300K? Such RLV providers would still be half as expensive as a sounding rocket (at least)? The reason for the $50K (or lower) price point, you will tell me, is to keep prices low enough to attract new customers. Exactly – low enough to attract new customers priced out of the current market. And going after new customers not served in the current market is called New-Market Disruptive Innovation.

    Full Circle. Thanks again for everyone’s ideas. Please do check out Ken’s posting on the Commercial Space Wiki site (see URL from Ken above). Ken has also posted some new articles based on Porter’s 5-Forces model that are excellent (and may be fodder for a future post.)


    -Colin

    ReplyDelete
  10. I spoke with Pat Bahn this morning about my presentation of this paper, and about a discussion he had with Clayton Christensen.

    His impression of the paper was "I agree with 85% of it and am in violent disagreement with 15% of it!" Actually, the point he made to me was that he doesn't think there's a low-cost disruption market because the current sounding rockets are virtually no-cost... it's the range costs that are the major expense, so no matter what vehicle is used, the end costs will be the same to all customers. He thinks the new-market disruption is what's coming, and he might just be right.

    Regarding his discussion about the emerging commercial space markets with Mr. Disruption Theory himself, after explaining the situation to Dr. Christensen, Pat said his (Clayton's) reaction was "This is gonna be HUGE!"... that's encouraging!

    Anyway, I'm going to send Pat the link to my paper and this page so he can get in on the discussion.

    Ken

    P.S. BTW, I've got a list of about 10 people now who are sufficiently interested in these markets and these methods... it's starting to become a cabal of sorts! ;-D

    ReplyDelete
  11. Anytime Clayton Christensen says, "This is gonna be HUGE!" I am encouraged!

    And thank Pat for getting your paper in front of Clayton. I love Cabals!

    ReplyDelete
  12. Colin,

    Just to make sure there's no misunderstanding... Pat spoke with Christensen a long time ago, long before my paper existed. That's what Pat told me Christensen said way back then!

    Ken

    ReplyDelete