Observations, signals and structures that liberate.

This post brings together some things I've been thinking about and using with groups who are working in collaborative structures or system change. One thing I often come across when working in this space is that groups find it difficult to create enough structure for the work without getting too bogged down in monitoring everything. I think it's partly because we've grown too accustomed to traditional monitoring, evaluation and learning techniques, theories of change and logic models. These have a place, but when much of your work is removed from directly delivery, they are often more harmful than useful. However it is important to have some structure, otherwise we often end up with inertia, or lack of action, because we aren't sure how to make decisions.

So, this is meant to be a guide to providing enough structure to learn and make progress. It's not magic. There are lots of other frameworks out there on systems change. Use what is useful to you to do the work.

Key terminology

Below is an outline of the terminology that will be used throughout the guide

	The mechanisms or methods you will use to affect the system (Convening & Connecting, Capacity building, Learning)
	The themes (worksteams) you are targeting (EDI, Wellbeing, Wealthbuilding)
	The levers within the system you are focusing on draw from your themes. These will be smaller parts of your themes.
	Observations: Things you are observing. This may be categorised as context, activity, process. These should be constantly collected from across the group and system. They don’t need to be overthunk.
	Signals: Things within the system you are noticing or expect to notice. Can be positive or negative, strong or weak. Some of these are likely to be related to your levers or at least have some connection. Repeated or related observations may indicate a signal.

Signals

Positive signals	may be signs of improvement or positive change. This could be something quantitative like improved wellbeing, jobs, funding.  could be less tangible like willingness to listen, approach to working, language used. Repeated or related observations may indicate a signal. may also be opportunities that could amplify the work, through collaboration.   may also be gaps where you can add value
Negative signals	may be signs of struggle or negative change. This could be something quantitative like wellbeing, jobs, funding. could be less tangible like an unwillingness to listen, approaches to working, language used. may also be where you are not adding value or actively conflicting with other work
Strong signals	Will likely be prominent, clearly observable patterns or indicators that suggest a significant shift in the system.  They will probably be well-documented, widely recognised, and carry enough evidence to indicate a high likelihood of affecting the system’s direction in the near term.  May require more urgent response or interventions because their potential impact is already visible and generally acknowledged.
Weak signals	likely subtle, emerging trends or early indicators may appear as minor anomalies, niche behaviours, or fledgling ideas and practices their immediate impact on the system might be unclear or small,  Weak signals are still important! They may require more nurturing.

Signals and Levers

It’s important to recognise that signals and levers are often related and interchangeable. You may identify something as a signal which is also a lever and vice versa. For example you may identify reducing poverty as a lever for change, but a reduction in poverty may also be a positive signal. Signals may become levers. There will be overlap, this is ok. 

Using Observations and Signals

Observations should be gathered from across the system constantly. Single observations may not feel important in the moment, but they may end up being hugely significant. When doing long term systems change it can be hard to notice change. Making observations unprecious can help to gather them in the moment.

There is no need to judge an observation at the time, just observe and record.

Having an approach to making sense of the observations and determining what they may mean, what signals they are indicating is important. Equally important is using observations for learning and adaptation. 

Don’t get caught up in the terminology too much. Just gather and explore. One way to think about the difference in their simplest terms:

• Observations are passive - they are things you notice
• Signals are more active - you are exploring what things mean

Observations

• Passive / Descriptive: They are what you notice, hear, or experience without needing interpretation.
• Raw Data or Context: Observations capture the immediate situation or environment (e.g., “The turnout for our meeting was lower than usual” or “We’re hearing more people talk about mental health issues”).
• Breadth Over Depth: Often a single observation on its own doesn’t carry significance yet it can contribute to a broader picture when combined with other observations.

Signals

• Active / Interpretative: Signals emerge when you begin to connect observations or interpret patterns.
• Indication of Meaning or Trend: A signal might hint at a shift in behaviour (e.g., “Multiple community groups are reporting a rise in mental health challenges, could this be a sign of systemic stress?”).
• Potential Trigger for Action: Because signals suggest a possible direction or change in the system, they often guide decisions, prompting deeper inquiry or immediate intervention

If we think of observations as related to activity, context and process we can easily use them to link to 3 questions

Activity - the work or things we are doing

Context - the environment we are working in

Process - the way we are doing things

• Are we doing things right?
• Are we doing the right things?
• How are we deciding what’s right?

 

Timeline and process

The diagram below is an example of a structure you might use. It is designed to build in regular rhythms. You should experiment with these rhythms and adapt

• Every 2 months - spend 10 - 15 minutes at meetings gathering observations. Use post'it notes and then move into Activity, Context, Process buckets. Even better if you can use a simple way for observations to be constantly gathered. You could use an online form, or even just a designated email address to send stuff to. Or just physically write it down (remember to keep it though!)
• Every 6 months - review observations and signals and explore the 3 key questions
• Adapt activity, levers and governance/decision making as appropriate

Balancing decision making and action

Decision making and governance are hugely important when working in this way. Governance can feel boring and/or difficult, but it is vital. How we make decisions is fundamental. Governance doesn't have to be how we've always done things, and especially in work like this, probably shouldn't.

There could be a whole post (or several!) just on governance, I'll probably follow up with one. For now I'd suggest focusing on getting some way of making decisions set up, so that at least you know what you are working with and what you might need to change. For now I've include a few slides with some simple things you might consider

Final thoughts

The most important thing out of all of this for me is to create a way to:

• make decisions
• learn
• adapt

There are many ways of doing these, this is one. Find a way that works for you