Sharing Lessons and Worst Practices

From KM4Dev Wiki
Jump to: navigation, search

Original Message

From: Charles Dhewa, posted on 2009/01/30

Hie Colleagues!

In the interest of learning and, with so many organisations learning from mistakes, would it be a good idea to document and share knowledge on Worst Practices?. We have had much about Best Practices.

Any views and experiences on this?


All replies in full are available in the discussion page. Contributions received with thanks from:

Frey Faust
Eva Schiffer
Amit Chkravarty
Laxmi Prasad Pant
Charles Dhewa
Ann lily uvero
Ram Babu
Paul Mundy
John Gray
Valerie Brown
Dr. Serafin D. Talisayon
Ssozi Javie
Amanuel Assefa
Stephan Ulrich
Pryor Tony
Nancy White
Matt Moore
Steve Song
Peter J. Bury
Deppendra Tandukar
Lucie Lamoureux

Related Discussions


The following high quality responses were triggered:

It’s a good idea but...

Ernst Bolliger:

I think, learning from mistakes is very common and human - nobody likes to hit his head twice in the same way. Learning from mistakes however necessitates a climate of trust. Without trust - and I think this is just a psychological phenomenon - we have much more interest to show our brilliant side. In our training and coaching events we always aim at learning from mistakes within the group: The first step is to build trust by declaring rules of the games (learning culture) and by honestly admitting own mistakes. However, by documenting and sharing worst practices on a open platform, I personally would fear soon to be known as "Mister Worst Practice" and thus loosing credibility. That is the hot spot: I do not tell the same stories on an internet platform as I do in a f2f group where I more realistically can assess the level of trust. It is a question of organisational culture how far I go with openly admitting failures and worst practice stories.

Laxmi Prasad Pant:

I love this discussion thread on whether we should learn from failures as well? In my own research, I am using two case studies - one being evaluated as a highly successful case and the other widely accepted as a dismal case of failure. My contention is that we can learn from both. But the BIG question here is this: are we comfortable to report our failure? Don't we prefer to cover up failures and report only successes? What is the risk of reporting failures? David Moss (2005)in his book "Cultivating Development" (Pluto Press)discusses the challenges of reporting/publishing critical stories upfront in the preface.

Amit Chakravarty:

I absolutely agree. We can learn more from our mistakes than from our successes. It is important to know "how not to do certain things". It saves valuable time and resources of future initiatives/projects. As already pointed out, nobody wants to discuss failures. Even if they are discussed the failures are sugar coated. I think rather than a bipolar classification of success or failure the focus should be on documenting and sharing learnings. Then the stigma of failure does not hurt.

Ram Babu:

It would be definitely interesting to learn from failure stories. Who will be willing to share experience of failure? Hardly someone. People may not be comfortable to report on failures because it may have several ramifications. This is human nature. In case studies, the failures or underperformance can be highlighted as a reference to facilitate the assessment of success. Case studies can provide information on what did not work, caused failures and what did worked out and overcame failures. Descriptions on the lessons learned also provide background information on what did not worked well and what needs to be improved or done in the future.

Dr. Serafin D. Talisayon:

Hi Ssozi, I agree with Amanuel that many people find that sharing "worst practice" is something difficult for them to do. I agree with you too, because not sharing worst or bad practice can be costly -- since others won't know what they don't know! I wrote about the cost of not sharing "bad/worst practice" on "Cost of Ignorance" in my KM blogsite last November 4 (click "F11- Cost of Ignorance" from here: <> Last October 28 in my KM blog I listed 8 KM mistakes I made. See: "Ooops! (Learn from my KM mistakes)" at

Ssozi Javie:

This is an interesting post/ idea. Worst Practices is the opposite of Best Practices. Sometimes it is better to use: "Don’t do this because if you do, the repercussions might be irreversible!" Rather than using: "do this and things will work out well." So sharing worst practices can be a good way of cautioning people resulting into best practices.

Eva Schiffer:

I think there is a general feeling that it can be difficult to share "worst practice" because it makes you look useless. But even if I overcome that fear, I think - especially if I share in a public forum - there is also the question of loyalty to everyone else involved in the failure. It's rare that I failed in a one-woman-show and while I might be comfortable with sharing problems, other people involved in the same project might think I'm betraying their trust.


Frey Faust:

Obviously positive approaches to learning are as important as learning from mistakes, one's own as well as those of others. In my profession, which is a lot about restructuring individual potentially movement habits, I and my trainees and colleagues profit immensely from the clinical and empirical research that has been done on past practices. Injury statistics are usually the starting point for many scientific studies, potentially the concrete negative effects of those practices.

Of course, the conclusions for preventative alterations can only be drawn if there is a consensus about what is healthy and this is a still expanding field. What I have noticed is that sometimes, even those practices which might cause harm have positive effects when they are used at the right time. Also, my students don’t necessarily need to have the overview of what might harm them all the time, because being overly aware of the dangers can inhibit, and inhibition brings with it its own dangers.

Ann lily uvero:

Very interesting discussions. Part of KM is learning from mistakes to improve a strategy. In one of the projects I was involved with, the field staff recommended 2 sites - one was success story and the other one was not. As part of the exit conference, the sharing of the CBO leaders who visited proved to be a learning to the CBO visited and vice-versa (what not to replicate). Lessons learned workshop, in the organizational level, with partner communities involved, not only organic/project staff proved to be a feedback in streamlining process and how the future program strategy should evolve. But interesting to note that the partner communities agreed that the PMU did 100% of effort regardless of the "mistakes". I also observed that admitting failures, is also cultural. One have to be careful in the facilitation process.

Paul Mundy:

Here’s a worst practice for you (names and places concealed so I don’t breach the terms of contract):

  1. The project is uneconomic. It involves building major infrastructure to benefit a relatively small group of farmers. The infrastructure is so expensive and the expected benefits so small that the benefits will never cover the costs of the infrastructure.
  2. The project is over-designed: it is complex and involves five groups of foreign and national consultants, two ministries, two project implementation units, and a central coordinating unit to try to hold it all together.
  3. Each of the foreign consultancy groups is required to report to a different ministry. These ministries do not talk to each other: the Ministry of Infrastructure regards the Ministry of Agriculture as amateurish and ideology-driven, while the Ministry of Agriculture sees the Ministry of Engineering as concerned only with pouring concrete rather than trying to better the lot of farmers.
  4. None of the foreign consultants speak the national language: not surprising since this is a country that has not ingratiated itself with western governments, so international consultants with in-country experience are scarce.
  5. Few of the staff in the project implementation units or the central coordinating unit are competent in their jobs: hiring policies in the ministries emphasize loyalty to the regime rather than technical competence. Nepotism plays an important role in hiring staff. Few speak English, so most can communicate with the foreign staff only through an interpreter.
  6. The foreign consultants have continual visa problems. Long-term staff have to leave the country every month or so to renew their visas. Short-term staff must await a visa approval before they can take on an assignment.
  7. All documents have to be translated into the local language or English. The project budget does cover translation costs. Automatic translation software is available, but the manager of the central coordination unit refuses to buy it.
  8. The project has a short-term international communication specialist (me!), who was hired to develop a communication strategy and do public relations. The specialist recognized the need to improve internal communication, and recommended developing a website to handle this. The purchase of the necessary software is still pending (and has been for 2 years). It is not possible to use free online services as these are blocked by the paranoid government, and do not have a national language interface.
  9. The international communication specialist had no national counterpart. When one was eventually hired, he resigned in frustration after a year. His replacement is inexperienced and is also responsible for monitoring and evaluation and management information systems.
  10. Most of the foreign consultants refuse to submit reports to the central coordination unit until they have been approved by their project implementation unit. This can take months. The project implementation units refuse to submit anything to be uploaded onto the website.
  11. The project has proved unattractive for foreign consultants and local staff alike. One team of consultants has had six team leaders in the first three years of the project. Getting a new consultant staff member approved can take 6 months or more. The central coordination unit has had five different managers, and the management of both project implementation units has changed several times.
  12. The consultants are supposed to develop systems and hand them over to local counterpart staff. These counterparts do not see it that way: they want the consultants to do the work, while they comment and criticize.
  13. The international communication specialist has discussed these problems at length with various managers, but they have not taken any action.
  14. The international donor has failed to recognize these problems or take remedial action. Donor visits have been frequent and disruptive, but have failed to solve any of these problems.

I could go on, but you get the idea.

Anyone got anything worse?

John Gray:

Why stop, Paul, just when we're all enjoying it?

Valerie Brown:

One group I work with has a great way of dealing with best and worst. After the workshop or paper, the members of the group each say what they liked/learnt from best and then how they each think it could be improved. This puts the onus on the critic to think of something better, not just a negative. We also have the rule of "critical loyalty" - so long as it is clear that the members closely support each other and share a common goal, all critcism is mutually helpful to reaching that goal.

For myself, worst things in my own social learning practice are:

  1. Thinking a new group is just like the last one, and so failing to wait to share their learning;
  2. Flashing out a comment when I think a member is trying to sabotage the group - I should let the group do that; and
  3. Relapsing into academic language; I know it is crucial in co-development of knowledge to talk clear, first principle language, but I revert to my old academic self.

Amanuel Assefa:

In one of our knowledge management projects, which was supported by Oxfam Netherlands in 2006, the key guiding principles to identify, document and share knowledge on sustainable agriculture and gender in Ethiopia were- to find out Good, Bad and New practices. We have noted that people were shy to respond/ share bad practices indeed. In this specific example, we have got reports on 36 good practices but only 5 were reported for bad practices. In the learning and sharing workshop, which was organized following completion of the data collection exercise, many people were not even happy with the naming- bad practices. The project has defined bad practice as any practice that is some how institutionalized and for which reasonably high amount of resources are spend but ended up to be less worthy in economic, social or environmental terms. In such cases the organization/person who is doing the practice will not repeat it in the same way it used to be, if he/she is given a second chance. In other words, bad practices do not refer to those practices at experimental stage (not institutionalized yet).

Any how some of the reflections and observations made by the workshop participants in relation to bad practice are:

  1. People are shy to report bad practices- because they may happen to be responsible for that (administrative reason) or have suppressive culture to talk about failures.
  2. Some people did not report bad practices because they were not aware of the fact that the practice they are dealing with is really bad. Such people may tend to be aware about the extent of the badness when ever they get chance to know more about good practices along the same line.
  3. In some cases bad practices were sources of innovation. It has been reported by some organization that having realized the bad practices they were dealing with, they have decided to spend lots of time, research, consultation etc so that they augment the problem-apparently leading to innovation.

Nancy White:

A dear and trusted facilitation colleague of mine introduced me to a method that allows a group to collectively explore worst practices. I have not tried it yet, but I am very interested in doing it. Essentially, the group is asked to imagine the worst possible outcome, then design a system to ensure that outcome. Then the group examines the elements of that system and reflects if any of those elements are in play in their own work.

The purpose of this process, Keith told me, was to recognize what we are doing that is not productive and which can be removed to be replaced by things that are productive. So it is not exactly telling failure stories or sharing worst practices but building an alternative scenario based on what we know doesn't work. Read the page below. I'm not doing a very good job explaining at 6:30 in the morning!! Fortunately, another group that used it with Keith made a video and shared a "how-to" page. If anyone tries it, I'd love to hear your stories back. selforganization.html

Eva Schiffer:

Dear Nancy,

That's an approach that really appeals to my German mentality - we always think through all kinds of possible disasters to be prepared. However, if you want to try this out in an intercultural setting, I'd be very careful. When I just arrived in Ghana and planned my first big conference, I had a culture clash experience with my Ghanaian colleagues about a very related issue. I asked them: What is the worst possible outcome and how can we prevent it. My colleagues refused to discuss that because they said: "If we focus on the negative here, we will get so de-motivated that we will loose all our enthusiasm and never even start doing it, because we might fail. We prefer planning with the best possible result in mind and dealing with the problems as they come along."

I think this points to a more general question of the cultural appropriateness of our tools and it would be great to utilize the cultural diversity of our community to get some feedback on how certain methods work in the different cultural settings.

Frey Faust:

I just wanted to share that I find a very tricky pitfall in our organization. Personnel preparation creates the need to consolidate education criteria, and this in turn formalizes the training and the training periods. Individualized approaches and fluid modification of systems adaptation stumble over these formalized criteria in two ways: one is that, when there is a consequent effort to update the approaches to more effective methods, those who were trained under older criteria are left behind or need to be updated, and two: the updating process causes delays and confusion as the older systems are re-vamped, because terminology has to be practiced and remembered before it can be deployed, because the reflex to do things in a certain order from habit causes hesitation, and hesitation brings a slew of minor mis-management events with it.

I am often faced with the prospect of over-turning an operating precept because of new information... and I hesitate often, knowing that we will be several months recovering.

Mario Marais:

Not sure this helps, but I've always been impressed by John Seely Brown of XEROX's distinction between organisational processes and the practices that make them work (In the book "The social life of information" if I recall). Maybe your operating precept is a combination of processes and practices, or it is actually at the practice level, at the essence level, which says how and why do we make this work for us? Processes are easy to change, practices not....and practices teach you how people actually make them work and what drives the work-arounds.

Frey Faust:

Thanks for responding Mario and the literature tip. I think process and practice are so closely wed in our work that it is as difficult to separate them as it is the seed from its destiny as a tree. Since our processing work is immediately translated into practice, and has a direct influence on the quality of the lives we touch through the information exchange, it is imperative to continuously update theory to fact and be on our guard about the unstable nature of some of these facts. In fact it is our responsibility to change our practices through the process of research, review, critique, and procedure updating.

I suppose you need a concrete instance... For example, clinical data strongly supports the idea that the Tibia rotates internally when the knee flexes. Often however, human motion training systems and knee flexing traditions do not reflect or respect this bio-mechanical report. Given the epidemic stats on knee OPs and replacements and other degrees of degeneration in teens to the elderly, it would be logical to assume that there might be a discrepancy between what we know and what we practice. If we, as movement educators do not update our information and chance our practice accordingly to include the clinically suggested biomechanics of the knee, we can also accept at least some of the blame for the injuries that result from the dutiful appropriation of our counsel and example. So this is the goad that forces us to progress... a good thing... but causes a bit of wasted energy... confusion and tends to exclusivity...

How we can deal with mistakes and bad practices


I absolutely agree that learning from mistakes/failures is a powerful way to learn. I've never, however, had much success in getting people to talk openly about failures, or produce case studies of failure. Even though many of us know the theory of learning from failure, there are too many deeply engrained habits of negative judgement around to make it easy to use that approach.

My guess is that learning from one's own mistakes is actually the most powerful thing - rather than learning from other people's mistakes - so perhaps the challenge is how to get self-reflexiveness into our ways of working. In the meantime, I'm thinking that it might be more useful to get people giving accounts of mistakes they've seen other people make, rather than their own! That would be less threatening. It's a bit like the research technique that has people being asked a question about 'other people like you' rather than directly about themselves when it's a sensitive subject.

Stephan Ulrich:

I think we can learn very different things from good vs. bad practices. It seems to me that every project or initiative that succeeds does so for a pretty much unique combination of factors - that is why it is so hard to replicate. Good practices usually inspire me to look out for factors and ways of doing things I had not thought of before. However, I am under the impression that failing initiatives mostly fail for the same reasons. Bad practice cases make me aware of a couple of critical factors, so that I become very conscious about avoiding these kinds of the big no-no's.

Thus, we need to have both. However, I agree that there are not enough bad practices around. But I don't think that publishing a bad practice reflects bad on oneself - as long as it is not the kind of "I was lazy" or "I squandered the money" bad practice.

Thanks for starting this debate, hope to hear more on this.

Pryor Tony:

I think Steve Denning’s point on this is very valid; you don’t often learn from a disaster, since you are running as fast as possible away from it, OR you are afraid that someone will overstate the problem and wipe out funding for an entire sector. And you don’t necessarily learn from a huge success, since you are so delighted that it’s hard not to just view it as a poster child for success, without stepping back and try to find whether the things that went right would go right anywhere, or were context-specific. Steve I think would argue that we usually learn from the good idea that doesn’t quite work up to full potential, the modest failure, etc. Many have already raised a key problem, though: how do you give people a comfort level to speak the truth, to lay out the weaknesses, without opening up oneself to unfair criticism from the outside. This can then lead ONLY to “best practices”, “success stories” and the like. This is not a trivial issue; both the World Bank and Company Command within the US Army have said that the best way to get a community to open up and be honest about what needs to be improved is to keep the door closed; limit sharing outside of that community. Of course, the response from outside, in the case of Company Command, is “who do you think you are? You are only a part of the picture; how can we improve the overall corporate entity, if you don’t want to share with other communities”?

A key to any of this is to have a decent understanding as to what was to be achieved in the first place, and to understand that in development few of the fundamental variables are within your full control. In some instances, failure occurs because “you got it wrong” but often it is because things just worked a bit differently than what we had assumed during design (possibly because something happened in between, a coup, a global recession, etc.). This then leads to a fundamental problem with case studies, evaluations, etc. For me the point is not to launch an unguided missile and then critique its flight path after the fact, but rather to learn, change, adapt and reconsider along the way. Except for the most egregious, I’d prefer to think of “failures and mistakes” instead as times when one can learn and adjust.

Matt Moore:

Not sure if this got mentioned in the conversation already but this is an example of learning from mistakes: . Worst practices are as important as good practices however there's a point to made about who we admit mistakes to and when. We are far more likely to talk about failure to people we trust in a private situation. I imagine different cultures also have different tolerances for when & where this happens.

Steve Song:

I think there is something to be said for "outing" failures. The trick is to do it with a bit of a sense of humour and a better idea. I just came across this site which seems to fit that bill.... See also

Nancy White:

This would be a GREAT topic for a KM4Dev online event!! I thinks this points to a more general question of the cultural appropriateness of our tools and it would be great to utilize the cultural diversity of our community to get some feedback on how certain methods work in the different cultural settings.

Peter J Bury:

Nancy, Absolutely an online event sounds great (anyway this is already becoming an event ;-). Let’s take it up in the discussions about forthcoming KM4Dev activities and events.

Charles, To you the question: how do you think the discussion is going? Is this what you are looking for? And most intriguingly too, what type of cases did you have in mind yourself when raising this issue?

To all, What to me would be most interesting is to hear from those (and some are around for sure) who were/are not on the 'offering' side but on the 'enduring' side!

Charles Dhewa:

This has turned out to be a great discussion than I expected. There have been many mistakes in the agricultural sector, especially here in developing countries and I was thinking policymakers and organisations would benefit if we share knowledge on worst practices.

I agree that most people find it difficult to admit where they have gone wrong. However, as KM4dev we could set a precedent by documenting some of these ideas and sharing them to, perhaps, minimize re-inventing of the wheels of mistakes.

Hege Hermansen:

I second an online event!

James J. Tarrant:

I agree. There’s a wonderful movement (still small) afoot that is trying to use GIS tools to do community mapping (usually of resource use – water, pastures, forests, etc.) with villagers in Africa, Latin America, etc. Under our FRAME project, we tried to support these NGOs and individual researchers and demonstrate these approaches through case studies, which were presented at a giant ESRI conference in Chile in 2006.

Meanwhile, more conventional community mapping techniques where you literally draw physical maps together with various types of forest communities (indigenous, migrant, etc) have a well-established provenance that is not often discussed here at least but which represent a really valuable KM tool. The maps can be (and have been) digitized and integrated into national GIS platforms in order to document forest use boundaries in situations where governments are deciding oil and gas concession permits (Peru). In general, I haven’t seen a whole lot of geographical or geo-spatial KM/KS techniques and approaches discussed in this otherwise superb community site. But, as someone who deals with environment and natural resources management issues in the context of development, geography and an understanding of traditional and modern use rights and tenure is so very important. GIS tools are already extremely important ranging from e-Government to natural resource and infrastructure planning. However, as noted above, the cutting edge is incorporating anthropologically rich data layers and sharing these with development folks designing and implementing projects. This same information will be increasingly important in designing vulnerability and adaptation strategies for climate change.

Deppendra Tandukar:

I would say, the mistakes should be documented as 'what did not work' but with all the context and situation at which it did not work. And, I believe this will give the alternative pathways to succeed the next time when you have to go through the similar project/path as you already know the pitfalls - you will be more careful and avoid those pitfalls.

Lucie Lamoureux:

Hi Charles, I was just about to write a message about documenting this, and all of the great threads that have been launched in the past week or so. For those who are new to KM4dev, we have been trying to capture our experiences in the Community Knowledge base:

The way we've approached it is by asking for KM4dev community volunteers to synthesize/summarize a thread in the Community Knowledge base so we can refer to it in the future. Any community member can volunteer, just ask me for a wiki username/password. Of course, it's very nice when the person who actually starts the thread by asking a question - and getting all those peals of wisdom ;-)) - gives back to the community by doing the summary... but this being an informal CoP, we can't force anyone to do so.

Anyone up for building our Knowledge base? :-)