Story:
MAKING OFFICE KNOWLEDGE OPEN SOURCE(ISH) - A CASE STUDY
How we identified a 10-15 % productivity improvement potential among knowledge and interaction workers attributed to a collaborative process and knowledge management solution.
This paper highlights some of the values that can be unlocked by using dedicated IT-tools and progressive management philosophies that enables and empowers employees of large organisations to control and/or influence a greater part of the processes and departments that they are a part of. The path that led to the conclusions is also described and some of the constraints and challenges are also discussed.
The problem
Problem introduction
Knowledge management is a rather multifaceted concept that can be (and usually is) interpreted in many different ways. The definition we stuck with is that Knowledge Management is all about: “making sure that everybody knows what and when to do that which is expected of them”. This definition is very easy to apply in a process driven industry setting.
Regarding knowledge management it is today a generally accepted fact that our brains should not focus on memorizing huge amounts of detailed data, simply because that is something our brains are not good at. Information needs to be stored however and the natural solution to that problem is best described metaphorically as writing books and building libraries, I say metaphorically because today this is done digitally.
The core problem
Many organisations (especially large industrial enterprises) are soo good at building digital libraries with information that it has spawned another problem: the digital library is so big it is difficult or sometimes near impossible to find the right “book” within a reasonable time frame (“book” = intranet page, shared folder, shared document, database entry, email, etc. etc.).
At a basic level of abstraction downstream this core problem the following symptoms can be identified:
- Long ramp-up times of newly hired talent.
- Long transition times for inexperienced employees to become experienced.
- Long internal adjustment times attributed to change efforts.
- High ratio of mistakes and internal quality errors.
To give a better understanding of this case it is necessary to understand what actually triggered the ambition to solve the problems above and where the ideas for the solution came from.
In early 2011 I was the “new guy” at an engineering department of a large industrial enterprise. To learn the ins and outs of my new department and get a clear and detailed understanding of how to solve the typical assignments I was going to face was quite frankly a mess. The number one strategy for knowledge transfer was to stand next to more experienced engineers and scribble down notes of all the details and eventualities as they happened. (Notes that, I soon learned, would be obsolete within a matter of years or even months).
Being part of the internet-generation as well as a dedicated DIY-enthusiast I knew that if all I need is the internet to teach myself how to build anything from tube-amplifiers and studio-speakers to plastic 3D-printers there must be a smarter way to teach the “new hires” at my company what they need to know to more quickly start performing at an acceptable level. Confronting my manager at the time with these thoughts ended up with me in charge of a project trying to find that better way.
Timeline & project management strategy
To truly maximize the potential for this paper to generate insight and inspire action the following paragraphs were added to cover the project management strategy and timeline in a slightly more detailed manner. The paragraphs are written in a very general way for 2 reasons, A) for the text to be applicable to as many scenarios as possible and B) to not cause any intellectual property and or confidentiality conflicts with my employer and other relevant stakeholders.
Many aspects of the project management strategy are best described as agile. The solution was introduced in the intended operational environment while it was being developed and user feedback and operational experiences where immediately considered when changes to the solution were designed and implemented. Also, we quickly saw that the topic of knowledge management carries a lot of stakeholders. To not fall in the trap of trying to fix everything for everyone we decided to aim for solving for the extreme in hopes of satisfying the in between. In our case the extreme was the typical newly hired talent, all other stakeholders were considered secondary.
It started out, however, as a typical pre-study and I was given a couple of hours per week to test & importantly further refine the hypothesis regarding “organisational improvement potential tied to improved knowledge management” which at the time was only backed by what’s best described as instincts and feelings.
Accordingly I set out to answer a number of qualitatively fundamental questions like:
- Is it a just a feeling or have we actually pinpointed serious (i.e. expensive) symptoms of (a) problem(s) that hence would be valuable solve?
- Does other parts of the organisation experience the same symptoms?
- Drill down, do the symptoms stem from the same root cause problem or a number of independent problems?
- Are the identified problems actually problems (i.e. something with a solution) or should they be seen as a part of reality that can’t be changed (without ridiculous effort)?
With a slightly more refined understanding of the case it was of course relevant to move on and answer questions like:
- Drill up, are there any other relevant but not yet considered symptoms of the identified root cause problem(s)?
- Have other departments tried to solve the problem(s) and/or managed to work around the symptoms in a satisfying way?
- Looking externally, is there a commercial “plug n’ play” solution available? What does academia and business literature say on the matter?
The pre-study resulted in a deeper understanding of the case and accordingly the hypothesis was refined to include some relatively specific suggestions regarding how to minimize the negative impact of the symptoms by attacking the root cause of the problem.
Translated to our case the symptoms were new hire ramp up time, adjustment times to change efforts etc. etc. and inefficient knowledge management was identified as the only “manageable root cause problem”. More details of the actual solution are given in the sub-chapter “Key Innovations”.
The pre-study results were strong enough to warrant a budget big enough to give the project a couple more hours of my time per week and, importantly, allow the project to hire a total of three master thesis students. The students were assigned to A) further refine the hypothesised solution and B) develop and implement an actual prototype web-platform that made up a crucial and very tangible part of the hypothesised solution.
Throughout the prototype development phase three main versions of the web-platform prototype were developed, and about one year after the first version of the web-platform was introduced in the operational environment of my department a series of interviews and survey investigations were conducted to validate the hypothesis and evaluate the concept solution and the project as a whole. The main results of this evaluation effort are described (in a general way) in the chapter “Benefits and Metrics”.
Today the implications of the results of this project are discussed and evaluated in the various instances that, in broader sense, could be affected of the solution (a process that can take years in an organisation as large as ours). At the same time actual use of the prototype web-platform and, importantly, the associated strategy is expanding globally to offices and departments tied to the same business unit as the one where the prototype was developed.
Key Innovations
The word innovation is a tricky word. It is typically interpreted in totally different ways by for example engineers, entrepreneurs, high level managers, investors patent lawyers etc. etc.. In the context of this paper, i.e regarding the project described here, focus is on the innovative application of existing or even rudimentary technologies and concepts primarily and secondarily on the development of new concepts and technologies. More specifically, the main takeaway of this project is what’s labeled in this paper as “making office knowledge open source(ich)” and any work done in the project that can be labeled as technically innovative has had the sole purpose of enabling that target.
Making office knowledge open source(ich)
The single most important project component was to design a process centric collaborative knowledge management strategy. As described in the previous chapters the main hypothesis of the project was from the very beginning inspired by the seemingly uncoordinated yet often efficient and, perhaps especially, very productive open source culture enabled by the internet and the web 2.0 revolution.
An important part of the strategy is to in effect delegate the burden of micro management to all relevant stakeholders. The word relevant being important and in essence meaning the people performing the tasks in question. This allows authority figures in the hierarchy to focus more wholeheartedly on large scale goals and problems that carries a higher value potential than micro management.
This strategy allows for a far more extensive “valuable control” in the sense that everything that is contributed to the shared knowledge of the organisation is made visible in all it’s relevant contexts and continuously evaluated and challenged by all stakeholders. In essence it could be described as evolution on steroids, any piece of information/knowledge that is shared but deviates from the optimal path to achieve a clearly set goal will be challenged (usually in incremental uncoordinated steps) leaving the information/knowledge representing the smartest/fastest way to achieve the goal as the official “office knowledge”.
Contextual filtering
In order to support this strategy a dedicated knowledge sharing and contextual information filtering system was developed.
When you start as a new hire somewhere you usually don’t even know what keywords to search for or what databases or folders to look in, hence a truly efficient solution must be “predictive”. In a sense this means that you begin your user experience at some generic starting position and the navigation options you are presented with are the ones most likely to be useful to you.
A predictive user experience can be achieved through extensive, automatic, user analysis and smart algorithms or syntaxes (like Google's predictive search tools) or by “hard coded design”. As creative engineers we of course have bias towards ideas including automation through advanced algorithms and syntaxes, but for our prototype, with its relatively small number of intended users, it was enough with a system based on “hard coded design” supported by smart syntaxes.
The contextual information filtering system we developed makes sure that every piece of information is displayed in a manner that reflects the bigger picture contexts. The main contexts that were taken into account are best described as the hierarchical context, the project context and last but not least the process context. Information hierarchically tied to one department is for example automatically linked with departments up- or downstream any shared "big picture process" (sales ⇒ quotation ⇒ manufacturing ⇒ logistics as a simple example).
It is still very important to point out that all these contexts are generated based on the data that the users contribute to the platform. It is not a consultant or administrator that visits all departments and creates everything for the intended users, the users themselves create the digital representation of all parts of the organisation that they in turn are a part of.
Getting too specific with examples and screenshots etc. can sometimes be dangerous because it is very hard to show the total depth of an idea (and we have to consider the confidentiality aspects etc.), but to not leave you the reader totally empty handed the picture below is attached. It depicts an early mockup of a page displaying an example department represented as an educational program in computer science, automatically associated with that department are the main routines that make up that program, classes, bureaucratic procedures etc.. Please don’t let this image limit your creativity (or your image of our creative capabilities), I dare say we are only scratching the surface of what’s possible.
For the purpose of this paper the three most important challenges specifically associated with the strategy and solution developed in the project are pinpointed below as the the 3 main requirements. Some more general experiences and insights gained in this project are presented in the chapter “Lessons”.
Trust: The nr one Requirement.
The suggested solution poses a problem if “the company” cannot trust that its employees are striving to achieve as good results as possible for the company. We decided early to take trust for granted because, frankly, it is not often that employees are actively working against their own company and if that is the case the company have bigger problems to solve before focusing on optimising knowledge management.
Control: The nr two Requirement
In hierarchical low trust cultures the right to interpret as well as the power to change the collective knowledge is guarded closely and kept within the “political territories” that reflects the hierarchy of the organisation. The most obvious symptom of this is high levels of micro management.
Still even if the trust levels are high and the official ambition is to adopt a more open strategy, large enterprises cannot give up the hierarchical, or chain of command, control over their projects, processes and structures all together and hope that the positive effects on innovation etc will end up being more valuable than the negative effects attributed to the potential chaos.
Hence the solution had to retain a very real level of control and authority figures and internal/external experts are still able to contribute with “surgical interventions” when necessary. (Our experiences are in fact that, utilised correctly, this solution allows such “step change” efforts to spread through the organisation much quicker/more efficiently.)
A comment on information security
Anyone working with knowledge or information management on a major organisation will probably shy away from the phrase: “make office knowledge open source” with very relevant complaints about information security. The philosophy of this project was always to create a culture that is as open as possible. But technically, the actual information shared on the platform is of course assigned certain security attributes requiring the proper clearance level prior to read or edit access.
Compatibility: The third requirement
The solution could not be just another information database system to add to the pile of of other “books” in the digital library, for this solution to be successful it had to be not only compatible with but actually encourage and assist better utilization of current “books”. (again, “books” = intranet pages, shared folders, database entries, emails, etc. etc.)
The values of using social technologies in enterprise settings is actually a well documented topic. One of the most exhaustive studies is “The social economy” released by the McKinsey Global Institute in July of 2012. A very important conclusion in that study is that accumulated over a range of activities social technologies in general have the potential improve the productivity of high skill knowledge & interaction workers by as much as 20-25%.
Those claims (and similar claims from other sources) were put into the perspective of our department and this project in the more refined version of the hypothesis that drove the entire project.
About one year after the concept solution was first introduced in the operational environment of my department a series of interviews and survey investigations were conducted to validate the hypothesis and evaluate the concept solution and the project as a whole. The main results of this evaluation effort are described (in a general way) in the sub-chapters below.
Short term values of collaborative process & knowledge management
We saw the MGI-study as a reliable source for our evaluation efforts and draw the conclusion that it is reasonable to assume that with our concept solution it is possible to realise at least 50% of the productivity potential claimed in that study. Important to note here is that the MGI study discusses the value tied to many different types of activities affected by to the use of social technologies, in our case we only looked at the the following activities: 1) e-mail handling, 2) Searching & gathering of information and 3) internal communication and collaboration. All in all the total short term productivity potential is about 5.5% or 8.7 hours per month and employee, see exhibit 1.
Long term values of collaborative process & knowledge management
The main purpose of this project was always to solve the problem of long ramp up times of newly hired talent. The evaluation indicates that improved knowledge and information management enables among other things a reduction in ramp up time of newly hired employees by 50% and internal adjustment time attributed to change efforts by about 35%. All in all the total long term productivity potential is about 7% or 14.3 hours per month and employee, see exhibit 2.
Easily shared insights
Influencing a conservative industry giant is quite a different challenge compared to for example shaping a new startup, but the values attributed to progressive management philosophies can be just as real.
If I were to list the three most important things required to achieve what we did it would be (in ranking order):
- You need a sponsor from high up in the hiearchy that believes in the project and the solution you try to accomplish.
- Involve users and stakeholders very early and listen to their feedback, their input is very valuable. Also, if you aim to change how people work this is much easier to achieve if everyone feel that they are included early in the process, it becomes "our" change ambition and not "your" ambition.
- Rely on people that are smarter than you when you develop the solution, it just gets so much better.
Another insight that could be worth sharing is that there is a tendency for very large organisations to “silo up” support functions and hence isolate them from front line/customer facing operational reality. There are many negative symptoms of this problem (or uninfluential part reality?) that goes both ways, but in this context I wanted to point out that this tendency carries with it a risk that the actual development of support systems (like strategies and IT-tools etc.) are A) predominantly seen as costs and accordingly something that should be minimized and B) turned into unnecessarily abstract and complicated challenges.
I am personally convinced that the reason this project could produce the (in my eyes great) results it did on the (shoe string) budget it was given is explained, in part, by the fact that I, i.e. the actual project leader, had very profound first hand experience of the problems we tried to solve. Hence for me, the level of abstraction was the lowest possible. (If a project is seen as cost or as an investment is very much tied to point 1 in the list above.)
And, to close the circle; The main purpose of the solution developed in this project is: To enable the individuals at lowest level of abstraction away from the “office knowledge of a department” to efficiently both read and contribute to the digital representation of their “office knowledge”.
The slightly longer discussion
We found this project very interesting and since it is very much inspired by the open source culture we decided to start a blog aiming to share our insights and hopefully inspire great discussions on the topics of enterprise 2.0, lean management, knowledge & process management etc. please check it out at www.papriun.com
The master thesis students that were involved in this project are Alejandro Garcia, Tove Attof and Oscar Björkman. Without their hard work and dedication this project would have never been as successful as it was.
You need to register in order to submit a comment.