KM on a dollar a day

Musing on knowledge management, aid and development with limited resources

Will I spoil KM if I tell people “best practices” don’t exist?

with 23 comments

Twice in the last week, I came across important internal guidance documents which mentioned the term “best practices” and the need for the organization to use them.

I was wondering, would  it be Scroogelike of me to tell them that just like with Santa, there isn’t really any such things as best practices.

(Aside: someone tweeted to me that they also have a problem with the term “practices” let alone “best practices”. I’m not going to take that on here – for me a practice is rather broad term for the sum total of the methodology, approach, set of actions and pattern of behaviours that we use in our work to tackle a particular problem – in other words it’s what we do at work. What is important is what we actually do “in practice” rather than whatever theory or idea we use to rationalize it).

So what’s my problem with the term “best practices”?

1. It implies we actually know the best way of doing something. In most cases we don’t. We might have developed some principles and good approaches based on experience – but this doesn’t necessarily mean that this is THE best way of tackling a particular issue. In some areas there might be a well established set of procedures, but these are usually best expressed in the form of guidance (and see my earlier critique of guidance). In practice though I’ve often seen the term used for practices about which there is relatively limited evidence or only a very few document examples of the practice being used. Often organizations (and individuals) are under pressure to quickly scale up successful approaches and this creates a risk of  over generalizing from a few successful examples. Also people who run successful programmes can overestimate how generalizable they are which leads us to:

2. We don’t always know WHY something works, even if it was successful. This is important because if we don’t know which parts of an approach are key to making it successful, then it can be hard to replicate into other contexts. Indeed it might be that certain elements of an approach can fairly easily be adapted from one circumstance to another while others are very context specific. If the context specific parts are also critical to its success then it can be extremely difficult to successfully transplant an approach from one context to another.

3. Use of this term can discourage us to look for ways of improving how we do things. All we need to do is follow the instructions, and so we don’t need to think too much about whether the approach makes sense in our current context. In fact we are led to the assumption that if things are not working then it is probably because we didn’t follow the practice properly rather than that the practice itself didn’t really work.

4. The description of a practice is not the same thing as the practice itself. However we document it and whatever evidence we collect, it represents only a summary of what happened and will most likely miss out important “tacit” elements of what made a practice successful, especially issues such as personal rapport between those involved, cultural sensitivities, the skills and instincts of those involved, as well as the importance of personal commitment to succeed (It is easy to include the need for personal commitment and ownership when documenting an approach – it is wholly another thing to reliably recreate it).

Despite these limitations I believe there is a value in documenting experiences and sharing them. To tackle this, in our work we have developed a set of three types of practices (and note the careful wording) to describe experiences that are promising (i.e. there may be elements that might potentially be replicated, or that important programming lessons can be gleaned from them for use in the future and in other situations):

Innovation – these are summaries of a programmatic or operational innovations that have or are being implemented under our mandate. These innovations may be pilot projects or new approaches to a standard programming model that can demonstrate initial results, and may generate potential learning for other situations.
Lesson learned – these are more detailed reflections (rather than just a description) on a particular programme or operation and extraction of lessons learned through its implementation. These lessons may be positive (successes) or negative (failures); both are valuable and encouraged. Lessons learned have undergone more of a review process than innovations, require some evidence to support them and generally have been implemented over a longer time frame.
Good practice – these are well documented and assessed programming practices that provide evidence of success/impact and which are valuable for replication, scaling up and further study. Documentation of good practices require more time and effort because of the need for assessment or evaluation results. The more evidence the better, as these practices should add value to development programming in a particular sector or region.



Each level requires a little more in the way of evidence of success and replicability (in fact we have more detailed descriptions of each including an indication of suitable information sources, and how these should be documented). In the case of “good practices”, of which we identify relatively few, we generally look to see if the same or similar approaches have been used in different countries and contexts, and how the results vary to help better understand what is “good” about the practices. Some colleagues had requested that we give detailed how-to guidance on the process for collecting these – but perhaps unsurprisingly we’ve resisted and rather chosen to share information on different methods that different offices have used, so they can see what works best for them.

A few important features of how these practices are defined and intended to be used:

1. They are not “final” but can and should evolve and be further developed overtime as we gain additional experience. What is a good practice now might be thought sub-par in future when we identify better ways of working, or the practice itself may be refined and improved over time.

2. They are not considered “definitive”. They are an example of a good approach not THE best approach. There may be several different good practices or approaches to dealing with the same issue. And different aspects of each of the practices might be combined or mashed-up to suit a particular situation.

3. They are intended to be used for inspiration, not for cloning. The idea is that those developing programmes can review the experiences of others and use them as a basis, or as a potential idea source for ways of tackling new programmes – but they should adapt them to their own circumstances, or choose to only adopt them in part or not at all based on their own judgement of the situation they are dealing with.

4. We recognize that the documented experience only captures part of the actual experience,  and that there is a lot of tacit knowledge in the heads of those people who worked on it from our side and the counterpart and partners side. We therefore encourage people to follow up and contact those involved in the original project to get greater insight into an experience they find promising and interested to replicate. In a sense the documented practice is intended to be the beginning of a process of thinking and consultation rather than the end of it.

This is not a perfect system, and we are constantly trying to improve it. We’re also trying to hold the middle ground by   explaining it to those who still think we need centrally validated “best practices”, and to those who think there is no benefit in documenting  practices at all.


Written by Ian Thorpe

December 7, 2010 at 10:51 am

23 Responses

Subscribe to comments with RSS.

  1. […] on a dollar a day has an interesting post today on “best practices“. I agree with the statement that a person (or organization) has to document lessons learned […]

  2. My understanding of best practices is that it’s relative, not absolute i.e. best practice in an industry is simply better than all the other ‘practices’.

    In the context of development, the term (in my experience) is used to draw attention to the often huge disparity between practices in different places but within the same country. So whilst practices in a certain industry of a developing country might on average be bad, they could be really very good in one part of the country and very bad in another part – suggesting that the mechanisms in place for facilitating dispersion of information need to be improved.


    December 8, 2010 at 6:07 am

    • Matt – thanks for your reply. For me the problem of “best” within an industry or country is who is to say whether it really is “best” and on what basis. Even within a single industry one practice does not work equally well in all situations.

      I fully agree on the need for improved mechanisms for sharing practices – but still that in each circumstance that project managers should make an (informed) choice on how to proceed to deal with the situation they face.

      Ian Thorpe

      December 8, 2010 at 6:30 am

  3. I understand your reasons for including the disclaimer about ‘best practices’ as a term.

    However, isn’t your key point that the term far too often is given meaning as best recipe, where it is much more useful to think of it as to date most successful way of approaching practice?

    Thereby making the argument exactly about ‘best practices’ as a discursive term.

    Sceptical Secondo

    December 9, 2010 at 3:51 am

  4. […] to matter a lot more. We too often forget that the theories, ideologies and grand approaches  (or best practices) are just models, they are not reality. They are only useful in so far as they are helping us move […]

  5. I find the notion of best practices particularly worrying when ‘international best practice’ is based on westernised norms that apply to that context, but the context where i am is totally different.

    just a simple example, many social protection schemes are in international best practice delivered at household level, but in SA due to a long history of apartheid-induced-migration the concept of household does not really apply — we live in a highly migrant environment, even 16 years after the end of apartheid.

    i know this is the case for some other southern African countries too — people move between households depending on particular financial circumstances at the time.

    the notion of a stable household is a western one. international best practice for social protection just doesn’t work too well in SA.

    Rebecca M.E. Pointer

    December 13, 2010 at 3:08 am

  6. […] Will I spoil KM [knowledge management] if I tell people “best practices” don’t exist? Ian Thorpe takes issue with the concept of “best practices”, and stakes out a better way for organizations to think about how they improve. […]

  7. […] the original post here: Will I spoil KM if I tell people “best practices” don’t exist? AKPC_IDS += "8181,";Popularity: 50% […]

  8. […] as regular readers of this blog will know, I’m not a fan of best practices. Rather we prefer to document and share real-life practices and experiences that can be used to […]

  9. Thank you for the take down on “Best Practice”. In my (limited) experience working in Pakistan (during the 2005 earthquake and then again with the floods) I saw that every few years there was a development catch phrase that was picked up and touted by both donors and recipients alike. I strongly feel that successful development models should be learnt from and adapted to the local environment. They cannot be replicated because what works one country may not and does not work for another. But the amount aid involved plays a very decisive role in that if access to donor money is centered around “best practice” then potential grantees will ensure that the either phrase, some random model or concept is mentioned in some capacity. Like they have to or else. I hate the term “capacity building.”


    January 7, 2011 at 12:58 pm

  10. […] and lessons learned from UNICEF programmes which can help inform and inspire (but not taken as a blueprint for!) programmes in other countries and other […]

  11. […] as regular readers of this blog will know, I’m not a fan of best practices. Rather we prefer to document and share real-life practices and experiences that can be used to […]

  12. […] as regular readers of this blog will know, I’m not a fan of best practices. Rather we prefer to document and share real-life practices and experiences that can be used to […]

  13. […] and living out loud“, “I want to be where everyone knows my name“, “Will I spoil it for you if I tell you best practice don’t exist“, and if only for the excuse to include a Buffy Video: “I’ve got a theory (of […]

  14. […] then everyone would know most of what they need to do a good job. I’m not convinced (see this past post on why I’m not a fan of “best […]

  15. […] presenta diversos problemas, entre los cuales encontramos claramente que los estándares (como tales) tienden a desprenderse de […]

  16. […] posts I’ve talked about the problems of relying too much on rules, guidelines and so-called “best practices” in development work. And while big organizations like the UN still rely too much on these, I’m […]

  17. Hi Ian,
    A few years late replying here, but what the heck….

    I’d love to completely agree with you – I usually do, but naggingly, I can’t this time. And it’s not because I believe in Santa.
    I think there’s a contextual filter that needs to be applied in order to establish whether good or best practice are relevant concepts.

    There will be some situations where best practice is a valid concept. I’d prefer to fly on a plane where the crew have been through a “best practice” pre-flight equipment check. I’d prefer to be operated on by a surgeon who had applied best practice in surgical preparation. And in the good old UK, I’d prefer to drive on the best practice left-hand side of the road! All pretty simple, operational, procedural practices – but ones where there is a recognized “best” which has become a standard.

    Here’s a timely one: If I was a golfer, then I’d use the best practice selection of a putter whenever I’m on a green.
    However – I wouldn’t view Tiger Woods’ golf swing as best practice. Or even Scott Adam’s one, come to that. Because a golf swing is contextual to the physique of the, erm, swinger. So each swing style is a good practice for the complicated combination of physique, age and experience. Hit a ball up a tree and you’re dealing with a complex situation where there might not even be a good practice for your Caddy to advise you on. Take an illegal drop, and things can get chaotic!

    So we’re taking a bit of a journey around the Cynefin model – Simple, Complicated, Complex, Chaotic.
    Where I’m sure we’ll agree, is that in the UN system, Complicated is about as simple as anything ever gets – so there is no such thing as an applicable “best practice”.
    Outside of the UN, in some more predictable, procedural environments, where there is, as you describes it in your guidelines post, a “best way” – so I would argue that there is a best practice.
    It’s just not in your world – which is no doubt what makes your job so much fun! 🙂


    April 15, 2013 at 12:48 pm

    • Hi Chris. Thanks for the feedback. Very timely as I was just making talking about this issue to a bunch of evaluators this morning! I guess I have two objections to “best practices” – one is the contextual one you mention i.e. that most of what we work on in the UN is complex or even chaotic and so “best practices” are not the right tool to use.

      But I also have another objection which is more semantic but I think equally important. You are right I’d want a surgeon who operated on me to be one who thoroughly knows established “best practice” in surgical preparation. But I’d much prefer it if these were the “best practices” of 2013 rather than those of 1913, or even 2012 for that matter. The point here is that “best” can give the impression of being static i.e. we know the best way to do something and so we don’t need to continue looking and learning. It really means is that it is the best way we know right now, but this, we hope, should change over time. For me the term “best” can imply that all we need to do is to follow protocol and everything will be as good as it can be – but even in complicated or simple domains of Cynefin I don’t believe this is always the case. Similarly while I want my aircraft pilot to know “best practice”, I also hope she will apply judgement and experience in applying those practices and if necessary be prepared to deviate from them, especially in the face of unexpected or unusual circumstances which may be rare but are neither unavoidable or fully knowable.

      So to stretch a metaphor – it’s good for people to be fully versed in current “best practices” where these can be identified – but I really don’t want people using this as an excuse for them to go into autopilot 🙂

      Ian Thorpe

      April 15, 2013 at 1:26 pm

      • Yes – I agree with you, whether it’s semantics or not. See, great minds *do* think alike!

        There is a real danger that best comes to mean “set in tablets of stone” and unassailable rather than best current practice. Mind you, some of those tablets of stone practices were pretty good…

        On a parallel, but related note – I’m involved in helping my church move it’s venue from a traditional church building to Ascot Racecourse – yes, I know, a little, er, unconventional, but we need the space, and it’s great neutral territory!
        Being a committed KM-er, I took my fellow church leaders to visit some other churches who meet on other racecourses to so some “learning before doing”. The comment made by one of our peers at Brighton struck me as relevant to this topic. When I asked him how long he thought they’d continue meeting on the racecourse site, he said “Oh, I don’t know – we keep that under review. We want to do whatever is best for the mission today.”

        Perhaps that’s best practice in a nutshell… best – for our mission – today.


        April 15, 2013 at 1:52 pm

  18. […] for a moment though, it reminded me of a recent discussion I had with Ian Thorpe on his blog entry “Will I spoil KM if I tell people that “best practices” don’t exist?”, which, incidentally has one of my most favourite ever Dilbert cartoons in […]

  19. […] as UN knowledge management and monitoring and evaluation specialist Ian Thorpe describes in his critique of best practices, following guidelines can in fact deter practitioners from thinking outside the box to look for […]

  20. […] that goes unused, or is used as a set of development recipes from donor to beneficiary (see this old blog of mine on why “best practice” databases are not the way to go), or a way of telling stories that make […]

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 )

Google+ photo

You are commenting using your Google+ 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

%d bloggers like this: