The Inversion of Knowledge

NotePeter Dorfman, former Knowledge Management Lead in the Office of the General Counsel at Hewlett Packard, has been a technology marketer, process consultant, author and thought leader in the KM field for more than 23 years. He can be reached at

The Inversion of Knowledge

By Peter Dorfman, © Copyright 2014, Peter Dorfman. All rights reserved.


Knowledge management was, in its early years, largely a top-down enterprise. Driven by a concern that corporate knowledge repositories would quickly fill up with inaccurate, useless junk without rigid quality review, organizations created small priesthoods of knowledge administrators responsible for virtually all authoring. Unfortunately, the results often were massive bottlenecks as content generated in this centralized way sat for weeks or months awaiting review. By the time knowledge reached its intended users, much of it had aged to the point of irrelevance.

Top-down knowledge management had limited success. It did not take long for proponents to arrive at the realization that KM would begin to show significant ROIs when the process was inverted. Centralized knowledge administration clearly produced higher-value knowledge — but centralized authoring retarded growth. Eventually, it began to be recognized, everyone — not just a small elite — would have to become responsible for generating the raw materials for corporate KM.

KM technology development followed the pattern. Adopters once favored highly engineered “expert systems” and similar artificial intelligence approaches for knowledge representation, until they experienced the difficulty and expense of implementing and maintaining structured knowledge bases. Structured approaches were in direct competition with more straightforward search-based tools. Search ultimately (and probably inevitably) won.

That was, of course, in the last millennium. Today, while there still are knowledge managers, knowledge engineers and knowledge bases, KM is an increasingly informal and an overwhelmingly social activity. Formal knowledge repositories are giving way to, or are supplemented by, informally-managed communities of shared interest. And managers are easing up on the expectation that content, once authored and subjected to elaborate review and refining, is then “finished.” Content is instead being deployed after relatively cursory editing and refined continuously through use.

This really is just KM finding its most effective orientation. Managers have found that allowing knowledge to be generated from the bottom up, through community interaction and a more informal knowledge capture process, wasn’t the compromise they assumed it was. Opening the knowledge capture process to more junior authors initially seemed risky – content that had not been created or extensively vetted by experts was getting into circulation and being used. But in important ways it was more realistic content, because it was based on actual experience, as opposed to being anticipated before it actually turned out to be relevant.

More importantly, bottom-up knowledge capture actually happened. Many organizations found that carefully designed top-down knowledge generation processes broke down, because the subject matter experts didn’t have the time to author. In addition, they quickly lost interest in authoring because they benefited little from the exercise. They were generating content for more junior people to use; it meant they got fewer interruptions to respond to questions, but the content was of little use to the SMEs.

Junior authors, on the other hand, wrote for each other’s benefit. They were both contributors and consumers of the content; in many organizations, it turned out they were more avid adopters of the KM process because of this. And the content created by junior authors was, to management, surprisingly good. Often, the quality of this content opened management’s eyes to the effectiveness of their own junior staff, whom they had assumed were not capable of generating effective material the way experts were expected to.

Today, when users of a product or service need answers to their questions, they frequently bypass “experts” entirely and just Google to see what their peers know. Often, the first place end users think to look for how-to advice is YouTube. Smart KM tool vendors are designing their offerings to accommodate this “self-help” tendency among their end users.

Knowledge management’s leading proponents saw this social element coming. In the customer support community, the developers of “Knowledge-Centered Support,” a set of KM principles championed by the Consortium for Service Innovation (San Carlos, CA), anticipated the migration of the KM discipline from a highly-structured and targeted knowledge base focus to increasing reliance on user community conversation (in KCS terminology, “from the funnel to the cloud”) well over a decade ago.

Personal experience in the KM field has reinforced this trend in my own practice. For example, while one multinational was developing its knowledge management practice in its global legal affairs function, it was simultaneously contracting its corporate law library down to bare bones, eliminating a once-highly-valued service through which the library provided custom searches of legal databases to its in-house attorneys. The downgrading of this service came in part because the company was getting similar search services from its outside counsel – but more importantly, executives in the General Counsel’s office realized that when lawyers needed information, they had stopped asking for help. They just Googled whatever they needed, and were quite satisfied with the results.

It has been observed widely that the ascendency of social knowledge sharing is a generational phenomenon – that “Generation Y” employees entered the workforce with social networking habits deeply ingrained and expected to indulge those habits in the workplace. While this probably is valid, the legal experience suggests to me that the move to the cloud is not as generational as the pundits assume – the Google-it-yourself approach was adopted just as avidly by the older attorneys as it was with the Gen-Y crowd.

What is going on in the cloud is the social element of KM, which in turn is an inevitable expression of the bottom-up evolution of the discipline. It is, on balance, a healthy trend.

Bottom-up knowledge generation will have significant impacts on the way work, and workers, are perceived by corporations. Management will have to develop new incentives for knowledge workers to contribute high-quality content. The most important element of this is time.

In order for staff to contribute effectively to a KM effort, they need assurance not only that management values this, but that they really have time to devote to it (i.e., during actual working hours). This is a difficult commitment for some organizations to make. Employees are expected to devote their time to prescribed work functions, and that time, and the work output, is measured. If those employees are told they are expected to contribute to a knowledge repository but not given protected time to do it, they will immediately object, because the work they are evaluated on will suffer, and this could affect their compensation. Some organizations offer their knowledge workers bonuses for knowledge contribution; a more effective incentive is protected time, not direct compensation.

9 thoughts on “The Inversion of Knowledge

  1. Pingback: The inversion of knowledge | Peter Dorfman Creative Services

  2. Although Peter and I loosely collaborated on KM at HP, this view of centralized knowledge administration, experts, and so-called junior authors is foreign to me. I am most familiar with the bottoms-up knowledge community at HP (formerly known as Hewlett-Packard). I played a central role in the community by founding HPedia, HP’s wiki encyclopedia. When HP acquired EDS, I worked side by side with the administrators of EDS’s wiki encyclopedia to merge the two. I left HP in 2010, but the last report I got from my HPedia colleague in 2013 said HPedia was “in good hands,” implying that it continues to thrive. All this is to reinforce Peter’s conclusions here: there is a powerful bottoms-up component to KM that can make knowledge sharing inside a corporation more effective.

  3. Pingback: The Inversion of Knowledge | Inventive, innovat...

  4. Pingback: The Inversion of Knowledge | Himanshu kakkar

  5. I think the article highlights two of the biggest barriers for significant bottom-up KM involvement. The first is the idea that contributions are not finished products but works-in-progress. What happens to these works if there is not a firm, continuous effort to add onto or revise? Does a KM system end up with half-finished or inaccurate reports? Is that a risk worth taking? The other is that the incentive for contribution as also noted in the reply by tbc0. Those at the bottom absolutely have to feel that it sanctioned by upper levels of management to spend work time adding to a KM system, whether it is formal or informal. Not only must be given the time, but they also have to be afforded the recognition that their contributions are of value.

  6. I think peer-to-peer knowledge transfer is an important part of a growing/thriving organization. KM systems that work need to adopt the behaviors people are already using.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s