In practice, Best Practice isn’t best
2 days ago
7 min read
Best Practice is one of those terms that seem to appear everywhere. In consulting proposals, presentations, workshops and discussions about how things should be done. I have used it myself often enough, probably without giving it much thought. It is an expression that everyone seems to understand and hardly anyone questions.
And I can see why. Nobody would seriously try to sell Average Practice. Even Good Practice sounds a little disappointing when you can offer the best. But the more I think about it, the more I find the term slightly amusing.
If everyone follows the same Best Practice, it may still be an excellent approach, but it is unlikely to make anyone particularly special. And if only a few leading organisations follow it, the interesting question is what makes it work for them, and whether the rest of us can reproduce those conditions.
Of course, this is partly playing with words. Best Practice doesn't necessarily mean the objectively best solution imaginable. Usually, it refers to an approach that has worked well elsewhere and is considered worth adopting. Fair enough. But the label often carries more authority than the evidence behind it deserves.
What are we actually buying?
We manage to describe surprisingly different things as Best Practice.
Consulting firms offer methodologies based on project experience. Software vendors provide predefined processes that fit their products. Industry organisations publish frameworks and guidelines, while benchmarking specialists identify practices associated with particularly successful companies.
All perfectly legitimate. But there is a considerable difference between a method that has demonstrated exceptional results and a standard process included in a software package. And yet, both arrive with the same label.
What I find interesting, is how rarely we seem to question the evidence behind it. What makes this approach superior? According to which criteria? Compared with what?
I also sometimes wonder how much of what we call Best Practice would survive a closer look. In less rigorous cases, the underlying evidence might amount to something like: I've seen this done before. It seemed to work quite well. Nobody complained, as far as I remember. Whether they're still doing it that way, I couldn't really tell you.
That may be a perfectly reasonable starting point for sharing experience. It is just a rather thin basis for claiming to have identified the best way of doing something.
Of course, there are organisations that systematically benchmark performance and investigate what makes certain approaches successful. Their work deserves to be distinguished from recommendations based primarily on familiarity or anecdotal experience. The question is whether we make that distinction when someone presents us with a Best Practice.
Best for whom?
Best Practices rarely come with all the circumstances that made them successful. An operating model that works brilliantly in one organisation may depend on a particular management culture, a level of employee autonomy or capabilities that simply don't exist elsewhere. The same structure or governance model can produce very different results under different conditions.
Yet the question of Best Practice sometimes comes up before we have even established what problem we are trying to solve.
Understanding why something worked is at least as important as knowing that it did.
I understand the motivation. Why spend time figuring something out if others have already solved it successfully? Learning from experience and reusing proven approaches makes perfect sense. But understanding why something worked is at least as important as knowing that it did.
There is also the question of what we are actually trying to optimise. Cost, speed, quality, flexibility, customer satisfaction? Improving one often comes at the expense of another. A practice can be excellent at achieving an objective that isn't particularly important to us.
The term Gold Standard, commonly used in medicine and science, offers an interesting comparison. It describes a recognised reference against which other approaches are evaluated, without necessarily suggesting that everyone should use it in everyday practice. A Gold Standard may be too expensive, complex or impractical for a particular application. Being the reference does not automatically make something the right choice.
And then there is time. Something that worked exceptionally well five years ago may have been overtaken by new technology, changing customer expectations or different market conditions. Even if we understand why it worked back then, we still need to ask whether those conditions exist today. For me, Best Practice becomes useful when we understand the thinking behind it well enough to decide what to adopt, what to adapt and what to leave behind.
When Best Practice becomes a limitation
Even if a Best Practice is genuinely good, well documented and applicable to our organisation, is following it necessarily what we want?
In many areas, the answer is yes. I don't see much value in developing a completely original approach to payroll processing, basic accounting or standard security controls. If there is a reliable, proven way of doing these things, I would be happy to use it. Being average in some areas is perfectly fine, particularly when the average is already pretty good.
And there are fields where spreading proven methods is precisely the point. In aviation, healthcare or industrial safety, we generally want effective safety procedures to become common practice. Their widespread adoption is a success.
But there are other areas where being like everyone else is precisely what we are trying to avoid.
Customer experience, product development, pricing, service delivery or even the way an organisation makes decisions can be important sources of competitive advantage. Adopting what successful competitors are already doing might help us catch up. Whether it will help us get ahead is another question.
Football offers a useful comparison. Teams study successful opponents, analyse their tactics and learn from their training methods. But simply copying the playing style of the current champion is unlikely to make another team the next champion. The copied approach may no longer offer the same advantage once everyone has learned how to respond to it.
Of course, doing things differently doesn't automatically make them better. There are plenty of expensive examples of organisations insisting on their own special processes when a perfectly adequate standard solution would have done the job.
The important decision is where we actually need to be different. That decision should come before asking which Best Practice to adopt.
Otherwise, we might end up implementing a very good solution to a problem we should have approached differently.
The comfort of following
Best Practice is also a rather convenient sales claim. It tells people what they want to hear. For the seller, it is an attractive proposition. We are offering something proven, established and apparently the best available approach. That sounds reassuring, particularly when the customer is about to invest considerable time and money.
For the buyer, it provides an equally convenient justification. Who is going to question a decision to buy the best? Even if things don't work out as expected, choosing an established Best Practice is easier to explain than taking a chance on something nobody has tried before.
Now imagine presenting an alternative: I have a rather unusual idea. Nobody has done it this way before, but I think we should give it a try. I suspect that proposal would trigger quite a few questions. And rightly so. New ideas deserve scrutiny, particularly when they involve significant investments or risks.
But shouldn't a recommendation labelled Best Practice face similar scrutiny? The label alone tells us remarkably little about the evidence behind it or its suitability for our situation. There is a related risk once the decision has been made. Implementing Best Practice can easily become an achievement in itself. The organisation has adopted the recognised approach, the project is completed and the decision no longer seems to require much justification.
Yet Best Practice doesn't replace continuous learning and improvement. Even a genuinely superior approach needs to be challenged as circumstances change. Otherwise, we may spend years faithfully following a practice that stopped being the best long ago.
Best Practice offers an attractive combination: a compelling promise for the seller, a defensible decision for the buyer and a reassuring sense of having done the right thing.
I don't think any of this is necessarily deliberate. But Best Practice offers an attractive combination: a compelling promise for the seller, a defensible decision for the buyer and a reassuring sense of having done the right thing.
The Best Practices nobody calls Best Practices
Not every Best Practice comes with a label. Many of the things we consider sensible or professional are approaches we have seen repeatedly, learned from others or simply become accustomed to. At some point, we stop asking why things are done a certain way. It is just how things are done.
I wonder whether AI might reinforce this tendency. Ask an LLM to design an operating model, propose a governance structure or develop a transformation roadmap, and you will often get something that looks remarkably familiar. Committees, roles and responsibilities, KPIs, reporting structures, phased implementation plans. Usually reasonable, nicely presented and sometimes surprisingly difficult to argue with.
Models can also challenge conventional thinking, evaluate alternatives and develop approaches tailored to specific circumstances. Much depends on what we ask and how critically we examine the answers. What changes is that a plausible recommendation can arrive without any explicit claim that it represents Best Practice.
With traditional Best Practices, at least someone makes that claim. We can ask where it comes from, what evidence supports it and whether it applies to our situation. With AI, the recommendation may simply appear as part of a convincing answer. Nobody has explicitly claimed that it is the best approach, yet we may treat it as though it were. We still need to question its assumptions and suitability, but there may be nobody behind the specific recommendation to explain or substantiate it.
Perhaps that is an even more convenient form of Best Practice: one that nobody has to sell, explain or prove.
In practice…
I am certainly not suggesting that we should stop using Best Practices. There is a lot to be gained from understanding what has worked elsewhere, avoiding mistakes others have already made and building on proven approaches.
But we should be more careful about what we mean by Best.
Sometimes it describes a genuinely superior approach, supported by evidence and measurable results. Sometimes it is a sensible industry standard. And sometimes it may simply be something that worked reasonably well in a previous project, at least as far as anyone remembers.
All three can be useful. They just shouldn't carry the same authority. The relevant questions are straightforward:
What has actually been demonstrated?
Under which conditions?
Does it fit our objectives?
And is this an area where we want to follow established approaches or deliberately do something different?
Comments