Much of my client work deals with learning, including the translation of data into learning and learning into practice, the facilitation of cross-sector learning, and the practices of becoming a learning organization. If you’re a regular reader, then you know these are also common topics in my writing. I’m a big fan of learning.
So keep that in mind when I say the following: Learning means a lot of different things, and you shouldn’t try to do all of them at once.
Taxonomy for organizational learning
In working with development and social change organizations, six types of learning turn up as the most important. I don’t always talk about them in this way with clients (better to use their language, rather than my own) but here’s what I’m usually keeping an eye on:
- Individual and team: Building the skills and knowledge of your team, including at both the individual level as well as the relational/joint skills of how a team works together (such as internal communication and team culture).
- Refinement-focused: Using monitoring or other data to understand your immediate impacts and what’s working, and then improving your approach based on those findings.
- Reflective mode: Taking broader stock of your approaches, opening the door to questioning your fundamental assumptions and radically changing your strategy or even your goals.
- Context sensing: Understanding your operating environment (e.g. political opposition, cultural norms, economic trends, weather patterns) and tracking changes that might have implications for how you work.
- Within context: How others (e.g. coalition partners, private sector actors, government agencies) are learning, even if that learning never makes it back to you.
- Cross-context: Learning across unconnected efforts, e.g. by using research findings from another country to inform your work, or learning in collaboration with those working elsewhere.
Two important things to note about this taxonomy. First, these categories are not mutually exclusive. Refinement-focused and reflective mode learning (often referred to as “single-loop” v. “double-loop” learning) feed into one another. Similarly, context sensing can help you make the most of cross-context learning, as you adapt external lessons to the challenges you face. Every social change effort involves a combination of these types of learning, often in overlapping ways.
Second, the relative priority of these types may vary greatly depending on the nature of your work, context, strategy, or team. For example, in relatively stable contexts, context sensing becomes less important. If you’re implementing a well established and highly evidence-based approach, cross-context learning may be very important at the start but less so as your work continues. With a more pioneering approach, reflective mode learning becomes more important.
What could go wrong? Timing and prioritizing
Many organizational leaders worry that they’re underutilizing all these types of learning—and some feel an urge to jump-start them all at once. I’ve often found myself saying: actually, you’re doing much of this better than you think you are. You just don’t always see it, either because it happens informally or because it doesn’t bubble up to the executive level.
Surfacing and formalizing those existing learning factors can help you augment them, but it can also undermine them. The idea of an OODA loop (observe-orient-decide-act) can help us see why that is. In general, you want that loop moving as quickly as possible, but you’re still subject to the rhythms of the world. Contrast:
- You might be able to iterate your outreach or advertising messages very quickly through some form of A/B testing, as your actions yield quick observations that can be easily processed into new orientations and decisions. That’s refinement-focused learning.
- On the other hand, the strategic impact of your work may take months to manifest into observable data, and distilling those observations into new orientations and decisions may require wrangling a number of stakeholders. That’s why reflective mode learning takes longer, regardless of how good your monitoring systems might be.
Trying to drive the cycles faster can undercut learning across multiple dimensions if you end up raising expectations and over-incentivizing certain types of learning. For example: Excessive data collection for refinement-focused or context sensing learning can lead to data fatigue (by those submitting data) and data drowning (by those trying to make sense of it). Individual and team learning takes time to percolate into programmatic changes; expecting short-term outcomes will undermine the case for long-term investments in this area.
Meanwhile, reflection workshops (for reflective mode learning) or commitments to externally sharable learning products (for cross-context learning) risk identifying lessons even where none exist. Few things are more frustrating than reading some lengthy research or learning report—always in an unwieldy PDF—only to realize that it adds nothing new to the general body of knowledge, and was only published to burnish the implementer’s brand or satisfy a grant requirement.
Cautioning against over-incentivizing learning seems odd when many leaders are more worried about the opposite problem, i.e. the active barriers to learning. If you’re having trouble getting started at all, then by all means: just start cranking away. But once you have a bit of speed and start getting over those barriers, you’ll find that generating new data, ideas, and insights is only helpful if you have an effective sorting/selection mechanism.
So if you want to better leverage learning in your work, think in terms of timing and prioritization. Which forms of learning will have the greatest impact on our work? Look for the bottlenecks and also the systemic leverage points where learning can lead to adaptation and greater impact.
Photo: Early morning cricket in western Nepal. Dave Algoso, 2017.