How to change the behaviour of your people: The end becomes the beginning

In this final part of the three-part feature, AstraZeneca’s Brian Murphy and The Creative Engagement Group’s Guy Champniss tell us how to move beyond the learning intervention and look to the long term.

Guy Champniss – The Behavioural Scientist

Measuring the behavioural change and sustaining the change for the future

It may be at the back-end of any behaviour change programme, but building robust and accurate measures for the new behaviours is of course critical. In tightly controlled experiments we may only have one measure (one dependent variable).

But within organisations we have noise; variables and influences we cannot directly control. To compensate for this, within field trials it’s important to be pragmatic and have a basket of measures. This way, if one proves to be unreliable or inconsistent, we at least know this is the case, through observing how it behaves alongside the other chosen measures.

We also need measures that are available in real-time. Whilst longitudinal data are important, with a behaviour change project we ideally want to get as clear a view as we can, as soon as we can. For this reason, we can turn to other measures.

For example, measuring attitude shift is a valid measure even for a behaviour change project (because attitudes are still a significant predictor of behaviour). Ideally though, we’d have a mix of measures – some self-reports (attitudinal) and others behavioural.

The advantage of the latter is they’re far less prone to social desirability bias (where individuals give the answers they think are wanted in some way).

With behavioural science, we’re confident we can tailor and target interventions to make sure learning becomes a repeating patterned behaviour that shapes and reflects the culture of the organisation. 

But any measure off the back of a behavioural intervention can only measure the effect of the intervention. Put another way, we may get the change we wanted to see, but is it going to last? Will it be a steady burn, or a bright phosphor burn and nothing more?

As behavioural scientists, we know maintaining a behaviour is a very different challenge to establishing a behaviour; each requires its own approach. And whilst building a habit can be one route to sustain learning behaviours, it’s by no means the only route. We can instead provide positive feedback, which tells the individual they’re getting better at learning.

Typically, we like doing things we’re good at, so the better we think we are, the more likely we are to keep going. It’s also the case that as we get better at it, we have to commit less effort (meaning lower costs in our cost-benefit analysis).


It could also be that individuals genuinely don’t realise the benefits of learning, and only by learning do these benefits become clearer and support new and stronger positive attitudes towards learning.

Linked to this, individuals may also learn something about their own tastes through the new behaviour. This concept of ‘taste discovery’ means that not only does learning provide new benefits, but it turns out it’s also pleasing to the individual on a more personal, intrinsic level.

We can also sustain the behaviour through making others’ actions visible; if we see others around us increasingly engaging in the new behaviour, then we’re more inclined to stay with the new behaviour too (especially if we’ve already committed, since to stop would mean a loss for us, which we know can be more painful than the corresponding gain in the first place). 

As mentioned at the beginning, behavioural science is pragmatic. As a discipline it’s committed to delivering the change needed. And nowhere is this pragmatism more evident than in the process to recognise that changing a behaviour is very different from sustaining a behaviour. We need different interventions for different moments in the process.

With behavioural science, we’re confident we can tailor and target interventions to make sure learning becomes a repeating patterned behaviour that shapes and reflects the culture of the organisation. 


Brian Murphy – The L&D Leader

Fuelling up the Rocketship

The great thing about applying behavioural science to help form an everyday learning habit is that it acts like rocket fuel for all future learning and change enablement. Digital disruption is fuelling the rate and type of change that we are facing in our organisations which in turn is leading to a greater need for innovation.

The forming of a continuous learning habit will therefore future-proof ourselves, our teams and our organisations.

It is the gift that keeps on giving and so getting good at it will be the differentiating factor for companies and their people in their quest to survive and thrive. Being comfortable with being uncomfortable and normalising this through incorporating learning habits as part of our every day will help us to create the learning agility that is needed.

Taking a data-first approach to developing a learning culture, particularly in organisations who increasingly are data driven by their nature, allows L&D teams to follow through on the performance consulting approach that I advocated for in a previous article.


Strong performance consulting calls for practitioners to start with impact measurement and sustainability in mind. Testing a behavioural hypothesis using evidence-based datasets helps to demonstrate performance impact – the holy grail for all L&D professionals.

But this isn’t enough.

Where this really gets impactful is when you can use the language of the business to tell stories that resonate and encourage a reinforcement of the change you are seeking to encourage.  Behavioural scientists call it ‘making others’ actions visible’, and is a form of social learning where the group will learn from one another and encourage collective learning and reinforcement. 

Build it and they will come

In order to sustain any learning habit (or any behaviour change for that matter) the environment needs to be shaped in tandem. Establishing the behaviour in the first place is only the start of the work. Sustaining and embedding this requires carefully orchestrated effort.

L&D become storytellers, community managers and curators, thus enabling the environmental queues to sustain the behaviour change that they are advocating. 

There must be environmental cues and signals that create the ‘pull’ for this new and changed behaviour. The L&D teams of the future must be environmental shapers, scaffolders of great learning and performance environments to help create the conditions for individuals and teams to ‘land’ their new every day learning approaches.

In the same way that cities with more green space, see higher levels of people taking physical exercise, L&D must help organisations create the pull for every day learning habits and propagate a true continuous learning and performance culture.

I’ve long advocated for a shift from courses to campaigns for corporate L&D teams, where the focus is less on managing learning but more on creating an employee-centric approach to workplace learning, supporting connection and community and enabling learning and performance improvement in the flow of work.

It signals permission and expectation that the organisation will support its people to own their own development. In this case, L&D become storytellers, community managers and curators, thus enabling the environmental queues to sustain the behaviour change that they are advocating.

There is a compounding effect of creating communities of early adopters to workplace learning, and then supporting this community to act as change-agents for the broader movement for change. This is a well-trodden path for OD professionals, it’s now time for L&D to follow suit.


The Learning Trial – how they did it

On the completion of the AZ Learning Trial, the L&D team found four key insights that really stood out.

INSIGHT #1 – Embrace the right behaviours

For the purposes of the Trial, the team identified three of the AstraZeneca corporate behaviours that best supported building a learning habit: curiosity, bravery and collaboration.

As a result of the Trial, they found that it is possible to significantly increase levels of bravery, curiosity and collaboration in approaches to learning. The results were statistically significant, meaning that the team could be extremely confident that the increases in behaviours within the intervention group were as a direct result of the interventions, and that they’d see the same effects across the rest of the AstraZeneca community. 

INSIGHT #2 – Seize the learning moments

The AZ team found convincing support within the Trial that increasing levels of bravery, curiosity and collaboration could lead to an increase in the ability to recognise learning moments day to day. Over 50% of the variability in this ability to recognise learning moments could be explained by the AstraZeneca behaviours of bravery, curiosity and collaboration.

By the end of the weekly behavioural interventions, they saw those in our intervention group as 64% more likely to be above the average of the control group. There’s a 60% chance that someone picked at random from our intervention group would be better at recognising learning moments than someone chosen at random from the control group.

INSIGHT #3 – Reflect and apply to make it real

The team at AstraZeneca believed that learning moments only come to fruition when participants reflect and apply what they learned to their everyday work. By embracing the right behaviours and noticing more learning moments the speed at which you put them into practice increases – this is learning agility.

In terms of a growing willingness to apply learning to our day to day jobs, the team saw that by the end of the Trial, those in the intervention group were significantly more willing than the control group. 70% of those in our intervention group were recording higher levels of willingness to apply than the average in our control group. And there was a 65% chance of someone picked at random from the intervention group showing higher willingness than a person picked at random from the control group.

INSIGHT #4 – Be consistent to make your mark

Murphy tells TJ “Learning isn’t an end. It’s a means to an end: and everyday learning can arm our people with the tools to make an even bigger impact at AstraZeneca, and for patients”.

As a result of the Trial, the team saw levels of self-belief to be able to make an impact increase significantly. Whereas self-belief to make an impact remained steady within the control group, for the intervention group, they saw steady rises across the six measures they were using to understand individuals’ self-belief. By the close of the Trial, 64% of the intervention group were above the average of the control group on this measure.

And compared to someone picked at random from our control group, there was a 60% chance that someone picked at random from the intervention group would have stronger self-belief in their potential to make an impact.

They also saw that a sense of personal impact was strongly influenced by an ability to recognise learning moments, and a willingness to apply this learning. 64% of the variability in one’s belief to have a positive impact at AstraZeneca was explained by an ability to find learning moments, and a willingness to apply.

Seize the learning moment

There’s an argument that learning and behaviour are intertwined, not just in the sense of how one approaches learning, but how one behaves to take advantage of moments to learn. 

Learning isn’t a passive process; it doesn’t happen to you. So, it’s critical to be able to actively seek out learning opportunities. During the Trial the AZ team found that bravery, curiosity and collaboration improved participants’ ability to find learning opportunities. 

Put another way, the behaviours had a major influence on respondents’ ability to recognise learning moments, their willingness to apply learnings, and their personal belief in their ability to make a positive impact.


About the authors

Brian Murphy is Global Head of Learning & Enterprise Capabilities at AstraZeneca and Guy Champniss is Head of Behavioural Science at The Creative Engagement Group


Learn More →

Leave a Reply

Your email address will not be published. Required fields are marked *