Deciding what makes a learning project complex, is quite complex

Mirjam Neelen

I’m working on a project to get to a mutual agreed LXD Process (at a global level) for learning projects that require a learning consultant. A question that came out of the project is: What type of learning projects require a learning consultant? One of the answers was that a learning consultant is required if it is a ‘complex learning project.’ This led to a discussion around ‘what makes a learning project complex?’

When I hear ‘complex learning project’, I think about complex skills. But I was pretty sure that wasn’t what my peers meant when they said ‘complex learning project’, so I decided to reach out to my network on LinkedIn to get some input:

Some people suggested that we should distinguish between ‘complicated’ (a lot of moving yet consistent parts, such as stakeholders, diverse audiences, multiple requirements, multiple technologies that need to be integrated, and so forth) and ‘complex’ (defined as ‘dealing with a lot ambiguity and/or change’, or even a piece of work that is not yet well understood, such as ‘we think we need to train people on X, but we don’t know exactly what needs to be addressed, how to do it, or to what extent – H/T Billy Wilson[1]), but for the purpose of ‘determining whether or not a learning consultant needs to be involved’, I don’t think that distinction is necessary.

Somewhat coincidentally (I’m saying ‘somewhat’ because Paul and I have been talking about complex learning and complex skills for years now), right before my ‘learning consultant discussion’ at work happened, Paul and I designed a learning journey together with Matthew Richter for the Learning & Development Conference (LDC; part of the Learning & Development Accelerator[2]), titled ‘Designing & implementing complex learning initiatives in organizations’.

In the learning journey at LDC, we defined a complex learning intervention at two different levels: The learning level and the systems level. In the learning journey, we used these levels to guide learners through tasks, such as task analysis, user research, designing for complex learning needs (all learning level), stakeholder management, politics, logistics (systems level), and so forth.

Next, I’ll use these two levels to categorise and describe variables that make a learning project complex. Just for clarity, what follows is my analysis and interpretation of what the ‘crowd’ said on LinkedIn. I’d also like to note that the majority of the comments were at a systems level.

Systems level

Any variable that refers to something that has to do with the ‘system’ in which the project takes place, is captured here. In other words, when I say ‘systems level’, I simply mean that it’s a complex project in an organisation that requires some kind of learning intervention.

VariableDescription
ChangeProjects that can be considered ‘change programmes’, with large audiences (20K +), that are diverse (e.g., multiple specialties, requirements, various tools and systems involved) and require a lot of change management. Rishibha gave this example, which we found quite shocking: “Skill building one group of workforce who will be taking transition from another group of workforce who will then be laid off after the skill transfer.”  
AmbiguityProjects that are ambiguous and/or unclear when it comes to: an internal agreement of readiness (e.g., re the change required) a common understanding of how fast you can/can’t move a common understanding around risks involved.  
Delivery  mechanismsProjects that have complex delivery mechanisms, such as multiple delivery modes, formats, methodologies, multichannel.  
Exploratory / Based on many unknownsThe full scope and substance are not known ahead of time. Sometimes, new ways of working and upskilling happen in parallel (this is risky and you need to make sure that you focus on prototyping and testing, iteratively). It could also be a first-of-a-kind performance or skills need for something that is not yet well-understood, for example “we know that we need to train people, but we don’t know exactly what needs to be addressed, how to do it or to what extent, and we may not yet even know how to determine if we’re hitting the mark.”   Note: I have seen a big increase of this type of projects over the last 3 years.
Impact measurement requiredProjects that require impact measurement add complexity, because getting the organisation aligned on metrics that matter and how a learning initiative can impact that, is not an easy task. It requires building a measurement strategy (and committing to it in terms of people and resources). H/T Josh Cardoz  
Note: (Competency) assessment and evaluation were also mentioned.
Audience variablesAny project with large audiences, with learners who vary widely when it comes to e.g., learning needs, work environment, location, technology capability and needs, ongoing support needs, and so forth.
Stakeholder variablesProjects with multiple stakeholders who each have different needs and demands, projects that require a steering group.
High riskProjects that have high risk, which can be, for example reputational, environmental, physical, financial, regulatory, people, etc etc.
TimeProjects that will run over multiple months or years.

Learning level

This one, I’ll tackle slightly differently. First, let’s define what we mean with ‘complex learning at a learning level’, using Van Merriënboer’s definition of complex learning. We need to do this in order to distinguish from the variety of learners’ needs I’ve listed at a systems level, because they don’t have to do with the complexity of the learning need in and of itself.

Van Merriënboer defines complex learning (see, for example this blog) as a combination of learning processes that happen simultaneously.

When it comes to learning, one way to look at complexity is based on the analysis of learning processes, where we can distinguish between schema construction (processes where people construct cognitive schemas) and schema automation (processes where people automate cognitive schemas).

When we look at schema construction, people can construct new schemas by connecting new information with what they already know, and they can also construct new schemas based on concrete experiences.

When it comes to schema automation, we’re dealing with rules formation and strengthening. Rules formation is about rules that are frequently used and then become cognitive rules. When these schemas are applied many, many times, so when there’s a lot of practice, these rules become more and more strengthened.

When all these processes occur simultaneously, that’s what we call complex learning. So, it’s a combination of different learning processes.

We can also distinguish between three types of learning, the integration of the following domains, which is a slightly simpler way to think about complex learning:

  • Declarative learning (knowledge)
  • Procedural learning (skills)
  • Affective learning (attitudes).

In the end, it comes down to answering the question: What does it take for people to apply what they need to learn in the actual work context? When untangle what it takes, the complexity will become clear very quickly (if you’re dealing with a complex learning need, that is).

On LinkedIn, in response to my question, people gave examples of this type of ‘complex learning needs’, such as:

  • Human Resources Information Systems training
  • Behavioral health specialist training
  • Systems training within variable work contexts
  • Projects where performance outputs, measures, or contexts vary

These examples are all examples of complex learning projects at a ‘learning level’, but unfortunately don’t necessarily help us to clearly see the variables that help determine if you’re dealing with a complex learning. Luckily, others, such as Van Merriënboer & Kirschner (2018) have clearly described the variables[3]. Paul and I have blogged about these before – here – and Paul wrote a whole book on complex learning, namely ‘Ten Steps to Complex Learning’ 😊. No harm in repeating them here, so let’s take a look.

VariableDescription
Multiple constituent skillsComplex learning needs require a combination and coordination of multiple constituent skills. The reason they’re called substituent skills and not ‘sub’ skills is because a complex skill is more than the sum of its parts. They’re are a bit like a human body, which is made up of organs. A body is more than just ‘a bag full of organs’. The organs must coordinate so they work well together. That’s also true for complex skills. When developing complex skills, you don’t just develop the individual constituent skills, you also need to learn how to combine and coordinate them, learning what to do when, how, and why that way.   Example: Driving a car includes monitoring other traffic, steering, checking mirrors, shifting gears, and so forth. All these skills are the constituents that make up the whole complex skill of driving a car.  
Integrated knowledge, skills, and attitudesComplex learning needs always require integrated knowledge, skills, and attitudes.   Example: Driving a car requires knowledge on how much pressure to put on the brake, depending on speed. It requires skills, such as monitoring other traffic, steering, and checking mirrors. It requires a ‘drive responsibly’ attitude. Knowledge, skills, and attitudes are all integrated when carrying out the complex skill of ‘driving a car’.
Controlled and automatic processesThe constituent skills involved in the complex learning need are performed in different ways. When acquired, some will be performed as ‘automatic’ processes, others as ‘controlled’ processes. Controlled processes require problem solving, critical thinking and decision-making.   Example: In the context of driving a car, breaking and using mirrors are examples of constituent skills that become automatic over time. When that happens, the learner can free up cognitive capacity to use for controlled processes, such as monitoring traffic in busy or unfamiliar places.  
TimeComplex learning needs require a lot of time to become ‘competent’.   Example: You might be able to get your driver’s license through a ‘crash course’, this doesn’t mean you’re a ‘competent driver’ at that point. You’re meeting minimally required standards that prove you’re able to drive safely. However, it takes years to become a competent driver (and some people never get there! 😀)

The people on LinkedIn also mentioned some additional variables that might add to the complexity of a project, for example:

  • Learning professionals not having the right skills – Many organisations have great learning professionals who are very good at things like content development, project management, programme management, and operational tasks. However, complex learning projects require other skills, such as performance consultants, learning design, learning technology, and data analytics. Organisations don’t always have learning professionals with these skill sets.
  • Deployment speed – This was mentioned a couple of times. Although we agree that short timelines increase project complexity, we’d also like to call out that these short timelines can be artificial and based on a ‘false sense of urgency’ as well. Keep that in mind!
  • Working with SMEs and/or Master Performers – One person mentioned that a variable that makes things complex is ‘when the experts themselves really struggle to define the rules they use’ (H/T David Gutiérrez García). This is a well-known phenomenon that indeed makes task analysis a complex task for learning professionals.

The crowdsourcing has definitely helped me to define variables to determine if a project requires a learning consultant. I also find it fascinating that learning professionals gave way more input at a systems level than at a learning level. I would love to hear your thoughts!


[1] Note that we’re not referencing ALL contributors; we mention people explicitly when we have copied/pasted some of their input in this blog.

[2] Disclaimer: Mirjam is a member of the Executive Advisory Board

[3] You can read more in this blog, or, if you want to dive really deep, you can take this FREE course on the basics of 4C/ID, with Mirjam as your instructor.

2 thoughts on “Deciding what makes a learning project complex, is quite complex

Leave a comment