…basically you throw everything into a pot and mix it all together into a smooth consistency. Then you stick a piece of bread on a fork, pop it into the mixture, and see what comes out. That’s your project learning, driven by a thinking process called retrospective coherence.

The Origins of Fondue and Project Reviews. If you are interested in Fondue have a look at this BBC Travel article, ‘Tracing Fondue’s Mysterious Origins’. For what I’m talking about here, all you need to remember is things get mixed together to create a gloopy (and delicious) liquid. What you probably don’t need to think about is the mixture coagulating in your stomach if you drink beer with it. Urrghh, that explains so much…

From a project review perspective I’m not sure when they became a ‘thing’. I don’t recall being involved in ‘project reviews’ and ‘post project learning’ in my early working life. It seems to have become standard practice since the concept of ‘Learning Organisations’ and PRINCE2 project management have become commonplace. What matters though is that they both exist.

Retrospective Coherence. There are two reasons for writing this post, Dave Snowden and ComplexWales. Both of them spoke about retrospective coherence within a couple of days of each other, or was it weeks? – see what I did there…

Basically retrospective coherence is a thinking process where we make sense of events that have happened in the past. By thinking retrospectively (looking backwards) we neatly put events in order, so that they make sense to ourselves and others when we ‘tell the story’ of what happened.

In this post The Mystical Acronomicon, ComplexWales explains how retrospective coherence often happens when people are writing case studies. The polished case study suggests a logical liner process with stages which neatly spell the acronym H.I.T.S. In reality however the stages happened as a convoluted mess spelling S.H.I.T. illustrated below.

For me this has a major implication for people using case studies. How can you replicate (copy) something if it didn’t actually happen the way it’s described?

From a project review perspective I think this case study example fits the Fondue metaphor perfectly – a gloopy mixture from which a precise example of learning is extracted (I’ll get my coat…).

The Dave Snowden prompt involved him speaking in this webinar about Risk and Resilience with Robert Koch from a power utility in South Africa. During the conversation they touched upon the process of extracting learning from incidents. Here are some of the key points I picked up on the need for rapid reviews: (apologies if I’ve misinterpreted any of these… it was a couple of weeks ago…)

• “you need to reflect upon an incident immediately – whilst the emotion is still with you”
• “the longer you leave it to reflect, the worse the information you get”
• “adrenaline changes our memory structures” (I need to understand this one a bit more)
• “put naive observers in alongside the incident team to capture the key learning points”

What I took from this was the idea that a project review held a significant time after the end of a incident isn’t going to yield learning points that are as useful as those that you get from capturing them in the moment (or as soon as you possibly can afterwards).

Rapid Reflection is Essential. I think this is a big deal. Lots of organisations invest heavily in project reviews and post project learning. I have encountered situations where people feel that nothing much happens as a result of them – the learning (if there is any) doesn’t get used and nothing much changes. Maybe they are just caught up in retrospective coherence and drawing out generic messages of little value from the gloopy fondue?

If you want to read more about retrospective coherence have a look at this post from Modus Cooperandi : Is Retrospective Coherence a Cognitive Bias?

Also of relevance, the transfer of learning in organisations is influenced by the rules and practices that exist – the learning we share depends on what the rules say. If you’ve done something brilliant, but it doesn’t fit with the existing rules, you probably won’t tell anyone. In this post I wrote about work of Professor Harry Daniels from Bath University; Field notes to report, what we write down in reports is different to what actually happens.

How do you make learning real time? This is a difficult one to answer. The easy thing to say is that you ‘make notes and keep a log as you go along’. I’ll stick that in the ‘bleeding obvious, easy to say, hard to do’ folder.

I’m sure that there are people and teams who will do this already. Digital equipment means that it is possible to record a vast range of activities. However, how much of the recorded material is used for learning and how much is actually for other purposes like protection against insurance claims? A kind of team ‘dash camera’ that only gets looked at if there is a crash or claim, never for routine learning?

Returning to the Cheese Fondue metaphor, this paper ‘Hot cheese: a processed Swiss cheese model’ in the Journal of the Royal College of Physicians of Edinburgh talks about the need for the continuous capture of learning. I’ve tried to capture the main points in this post, Struggling with Learning from Failure? Just host a Cheese Fondue Party which ends by making the point that it’s better to try and learn as you go along, rather than leaving it for the Fondue Party at the end.

So, What’s the PONT?

1. Learning from project reviews is a good thing. Just be aware of the pitfalls particularly if you leave it for a long time.
2. Try to capture learning points in real time or as soon as possible – when you can still feel the emotion.
3. Think about involving people as ‘observers’ to stand on the ‘outside’ of the activity and record the points of interest and potential learning .