BP’s spill & KM excellence: A paradox?

BP’s spill & KM excellence: A paradox?
June 3, 2010 Alice MacGillivray

Photographer Brian Skerry tells us that 90% of the big fish in the ocean have disappeared in the last 50-60 years. And most of us didn’t even have that context as we turned our eyes to the BP oil disaster in the Gulf of Mexico. It’s early June and there has already been a MEDEVAC of a worker with chemical poisoning. The leak may not be fully stopped until August. The effects of chemical poisoning in a human—according to CNN news—can last for more than a decade. We have no idea about the massive number of non-human lives lost. Oil and dispersal chemicals will be caught up in currents and spread well beyond the Gulf. And meteorologists are predicting a bad hurricane season.

Those of us who work with Knowledge Management (KM) know that BP has—or had—an exceptionally good reputation for work with knowledge as an asset. BP won the Most Admired Knowledge Enterprise award many times, and many KM publications draw from the years when Lord Browne was CEO. How can a company that developed such a high skill level be at the helm of such a disaster, apparently uncertain about everything from technical solutions to communication?

I can think of many hypotheses related to knowledge management and they are just that; I have no evidence. One has to do with the nature of the KM processes and tools BP developed and used, and those they didn’t.

I heard one of BP’s well-known KM experts present in 2003. He cited Lord Browne’s catalyzing quote: “Anyone in the organisation who is not directly accountable for making a profit should be involved in creating and distributing knowledge that the company can use to make a profit.” (In this post, I set aside the for-profit element). After the presentation I noted that almost all the focus had been on knowledge sharing and distribution through communities of practice, virtual teams, action reviews, and so on. I asked about the “creating” part of Browne’s knowledge equation. (At the time, we were very much aware of the limitations of knowledge sharing or transfer; it is part of the reason we have retained that awkward umbrella term of “knowledge management.”) Knowledge generation was not being ignored—he explained. It was an integral part of the peer assist process.

For those not involved with knowledge management, a peer assist is used to improve plans before they are put into action. Typically a team that has successfully completed a major project flies in to meet with a team about to undertake a similar project. By exploring the rich edge between experiential learning in another context and knowledge of the new context, valuable new knowledge can be created.

Yes, this is a great example of knowledge generation. But it is tightly bounded by the project mindset. Drilling for oil is a standard part of BP’s suite of activities. Capping a leak that could never happen from a rig that could never sink is not. Might the epistemic culture of engineering (would I be accurate in saying this kind of engineering is best with how, what and when questions?) restrict the possibilities for exploring different kinds of questions requiring knowledge generation?

Comments (5)

  1. nick m ilton 14 years ago

    Alice, I am replying as an individual with a very strong background in KM at BP, but answering as an individual and not in any official capacity. I am not a BP employee any more.

    A couple of things here.

    One is that the best KM systems cannot protect us from the unknown unknowns – the Black Swans. What we appear at first sight to have here, is a failure of one of the most robust and tried and tested pieces of equipment in the oil sector – the Blow-Out protector. Why it failed, we dont yet know. We will know after the enquiry, and then we will understand more. Hence I think there is no paradox.

    Secondly, my personal understanding is that BP is not “uncertain about …… technical solutions”. There is a repertoire of technical solutions, but all are extremely difficult at these depths. The technical solutions are actually quite simple in concept, but very difficult to deploy in this circumstance. They are also seeking knowledge and suggestions from others, and evaluating these, so we will see if there is something “out of the box” that can be done, but I expect the successful solution will be something already known, but just very difficult to apply in this context.

    You mention Peer Assist. Peer Assist often draws in people from outside the company to bring in out-of-the-box learning. In these cirtcumstances it is a knowledge creation exercise; bringing together disparate bacgrounds in the context of a discrete problem.

    All the best


    • Author
      amacgillivray 14 years ago

      Thanks so much Nick; I’ve been hoping to gain some inside perspectives, and perspectives from someone no longer in a system can be additionally insightful. I’d like to pose a couple of thoughts here, hoping you’re willing to share more insights. Not knowing the sector, I may not be phrasing questions properly; feel free to adjust them.

      You talk about the Blow-Out protector being one of the most tested pieces of equipment in the sector. Am I correct that testing of this and other equipment can be extremely expensive? What drives that type of fiscal commitment in a company such as BP? I.e., how wide is the boundary? Profit? Reputation with shareholders for profit? Environmental protection to avoid fines and loss of reputational capital? Concern for the environment and citizen’s health as something that might decrease profit but is important?

      Coming back to my sense that [some branches of] engineering appear to be “How” “What” and “When” oriented, it seems to me that there is a potential paradox between testing something rigorously and identifying black swans. In the absence of a “what if” focus–for example–as something is tested and tried, it gains a kind of invulnerable status. The more testing, the greater that perception, and the greater the likelihood of thinking of failure as a black swan. Does this make sense? But I think you’re also suggesting this wasn’t really a black swan, in that “the technical solutions are actually quite simple in concept.” So, perhaps what ifs have been well explored, with tested solutions in place.

      Yet, the tested solutions aren’t working in a real (and therefore planned) context. Does this come back to the cost of testing?

      On another note, I always emphasize the knowledge creation elements of peer assists in my work; good to hear your support for that and the point about bringing in “outsiders”.

  2. roanyong 14 years ago

    Alice, I agree with Nick that there is no paradox here. BP excels in its well-known Peer Assist (PA) program. However, the assumption underlying the PA is somebody out there (whether they are inside or outside the organisation) have the solution. Alas, in the oil spill incident, there is no one who knows how to solve the problem.

    I respectfully disagree with Nick however that ‘Black Swan’ is the cause of this unprecedented disaster. In my view, ‘Black Swan’ is an event that nobody can think of happening. For e.g. the 911, or the recent Icelandic volcano Ash. The oil spill, however, is not ‘Black Swan’ event because as early as 2003, some people have voiced concern over a possibility of un-manageable oil spill. The warning was tossed aside by the BP executives, because they thought the Blowout preventer was sufficient.

    That’s why Snowden’s Cynefin framework is useful to analyse BP’s situation. Peer Assist and Best Practice were the best techniques to solve ‘simple’ problem. In this regard, BP is the role model. Unfortunately, the oilspill incident is in the ‘complex’ region, which mean at least two things:
    1. Experts could not solve this problem
    2. The engineers and experts need to continue probing the problem until they can make sense of the problem, before they respond.

    For more explanation, I’d invite interested readers to read my blog post:

  3. Author
    amacgillivray 14 years ago

    Well, this isn’t technically a reaction to my post, but it shows other reactions to the issues I raised.
    See Tom Davenport on “If Only BP Knew Now What it Knew Then” (HBR) http://blogs.hbr.org/davenport/2010/08/if_only_bp_knew_now_what_it_kn.html


  1. […] received from Nick Milton and Roan Yong on my post “BP’s spill and KM excellence: A paradox?” Both–in somewhat different ways–challenged the idea that there was a paradox: […]

Leave a reply

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