Build your change approach using this proven technique that Mckinsey consultants use

Build your change approach using this proven technique that Mckinsey consultants use

Often change approaches are built not using a structured and well-defined series of logics but often using a cookie-cutter, standard change approach or change strategy template.  In some cases, a ‘gut-feel’ may also be used based on stakeholder wishes or preferences.  These are subject to stakeholder criticism nor do not achieve the goals of the initiative.

The usual standard change approach for most initiatives often entail the following.

  1. Stakeholder engagement – setting up forums and sending emails to stakeholders
  2. Training – planning for and executing training delivery to ensure users understand the new system/process
  3. Pre-go live readiness – send reminder emails and build intranet articles to raise awareness

How might we better derive change approach using sound logic and an evidence-backed approach?

One way to construct a logical, structured and well-thought-through change approach is by using a logic tree.  A logic tree is a visualization that captures all the component parts of a problem, in order to make it easier to identify a hypothesis that can then be tested using data and analysis.  Logic trees are great for making decisions by visually decomposing the various elements and reviewing these holistically. 

In the following example, a family uses a logic tree to decide which new town to move to by narrowing almost 30 possible potential locations to just one.  In the following diagram (from Bulletproof problem solving, Wiley 2019) you can see how this family started with the problem it is trying to solve, and then broke down the problem into its elements.  Then within each branch a weighting is assigned to each branch, in terms of percentages.  Then each sub-branch is also assigned weightings. 

Example of a logic tree in deciding where to live

Then as a next step data can be collected to determine which town meets the various criteria as defined in the logic tree.  By doing this, laying out the various components, and analysing its weightings, you can derive the best location.

This is how McKinsey consultants and other strategy consultants solve large complex problems.  The logic tree forces you to structure your problem versus being lost in focusing on certain approaches and neglecting others.  Any problem can be solved using this approach.  Even the largest of problems can be broken down into its smallest components.

Strategy consultants then go through every branch to analyse them and collect data to prove or disprove each branch one by one.  This means, that each branch or hypothesis is tested and proven or disproven.  In this way, every option is considered and the chance of making a wrong decision is greatly reduced.

So how might we build a logic tree that helps derive the change approach for an initiative?

  1.  Start by defining the problem or question to be addressed.  What is the goal of the initiative?  Is it to implement a new system that is fully adopted by its users?  Is it to increase cross-selling by sharing customer information across business units?
  2. Think of the broader buckets of each branch.  What are the core types of change approaches to address the problem?  Think widely and carefully about all the types of buckets possible that would address the problem.
  3. Expand the branches until you have covered all possibilities
  4. Go through and assign a weighting in percentage terms to each branch and then use this to determine the focus and importance you may want to place on certain branches in terms of research and data collection
  5. Go through each branch and systematically to reject any that do not apply based on data.  For example, one branch could be to use video as a channel to communicate.  However if the data shows that previous usage of video to communicate key messages did not result in raising awareness for this stakeholder group, then reject this option

Here is one partial example of deriving a change approach for a customer complaints project.

Logic tree example of a detailed change approach

Click here to download this diagram.

One important principle to note when building branches is to ensure that the branches are MECE.  MECE stands for mutually exclusive and collectively exhaustive.  Mutually exclusive means that you need to ensure that each branch is unique and does not overlap with any other branch.  Collectively exhaustive means that every option or alternative has been considered and nothing is left off.  In this way, the branches you have built are bulletproof from a logical structure perspective.

Building a change approach using a structured approach that is data-supported and logic tested will earn significant stride with the most critical of your stakeholders.  You can even hold a workshop to work with your stakeholders to define the logic tree and assign weightings so that that the agreed approach is one that is clearly visible and logically sound.

Another important point to keep in mind is how each of the branches of the logic tree change approach will interface into the overall change environment. When planning on the execution of the overall change approach or each branch of the approach, one needs to be clear around the velocity and volume of change and what else is happening in the change landscape. Using data visualisation tools such as The Change Compass is one way to grasp and plan around the change environment.

Top 7 challenges faced by change practioners in generating insights from change data

Top 7 challenges faced by change practioners in generating insights from change data

We surveyed senior change practitioners on their key challenges in using change data to generate insights, and here is what we found …

Change practitioners seem to face quite a lot of challenges across the board in measuring change and demonstrating the value of managing change.  For many, there appears to be a level of angst and frustration in not being able to break through and demonstrate insight through change data in a clear and simple way.

Why did we survey this topic?  In the new economy, our world is increasingly dominated by technology and data.  More than ever data is all around us and our ability to access a range of data is becoming more prevalent.  At our fingertips, we can access our phone to see how many steps we have taken today, work email and even workplace chat platforms.

In the business world, the same applies even more so.  All facets of how business is being will increasingly be dominated by data.  The availability of data.  The insight that can be generated by data to make decisions.  Data is king and a competitive advantage.

However, in change management, our ability to use data has mainly been restricted to ‘soft’ qualitative data.  Of course, all types of data are useful both hard and soft data.  However, most of our stakeholders who make decisions on project execution, funding, and prioritisation are focused on hard metrics.  We really cannot blame them because hard metrics tell a direct compelling picture, whereas soft, qualitative data requires a level of interpretation and maybe less direct in the implication.

We surveyed a sample of senior change practitioners and received 30+ responses.  After sorting through the feedback and responses we grouped them into the following 7 themes.  We also directly address each of the challenges posed.

1.  Getting buy-in from stakeholders on data input

Some mention the importance of stakeholder support and buy-in in collecting change data.  This can be quite challenging if your stakeholder does not see the value of the change data that you are collecting.  Since the bulk of change data is derived from each of the impacted businesses and those involved in the initiative, it is critical that the impacted stakeholders are supportive to ensure that data may be collected and response is sufficient.

A key element in ensuring that your stakeholder buy-in to your change data plans is to come up with a ‘sales pitch’ for them personally.  Each stakeholder is concerned about their own priorities and challenges.  If the change data can be positioned to address one of their pain points, then it will be hard to imagine any stakeholder who will not be interested.

2.  How to measure cultural & behavioural change

This is probably the biggest challenge called out across respondents.  Most change practitioners work on embedding some form of behaviour change.  As a result, being able to measure the behaviour change is critical to demonstrate the value of having a change manager onboard and the value of change tactics. 

And since most initiatives are not end-to-end transformations of everything within the organization, there is usually a limited set of behaviours that the initiative aims to change.  Working on measuring a small set of behaviours can be challenging because it is not that we are measuring the whole culture of the organisation, which can be measured by culture inventories such as Organization Culture Inventory (OCI).

One way to do this is to start by defining the actual behaviours you are trying to measure in very specific detail, in a way that is behavioural and observable.  For example, customer service representatives will be able to resolve customer complaints in the first contact without escalating to their team leader.  This can easily be measured using the data from the CRM system that the representative uses. 

Then we can break this down into more discrete ‘micro-behaviours’ that will contribute to the overall behavioural outcome.  For example, in this example, it could be 1) Establish rapport within the first 3-5 minutes of the conversation and 2) ability to identify a customer complaint 3) Apply structured complaint resolution strategies as per training content 4) Regular supervisor coaching and guidance on complaint resolution performance.  These behaviours can be recorded using call listening audits, self-ratings, and/or supervisor ratings.

3.  Data requires time, resources and effort to collect. 

Change practitioners told us that the amount of work involved in collecting, sorting through, and analysing data is very resource-intensive.  Because of this many try and avoid this as they do not have sufficient time or resources to collect data.

A lot of the work required is also very manual.  Many mentioned automation as something they are looking forward to.  Change data that can be automated to save time and energy to follow up, collect, followed by data analysis is one that everyone looks forward to.

The solution is to leverage various digital tools to better automate the capturing, analysis and visualization of data.  For example, Change Tracking is a tool now owned by Accenture that measures change readiness and generated reports.  For various task management and collaboration features, most use such as Trello or Jira/Confluence.  To measure change impact and change capacity, try The Change Compass.

4.  Change capacity

The capacity for the impacted business stakeholder to undergo and embed the change is often the first that comes to mind when it comes to change data and reporting.  Most respondents mention manually developing a change heatmap to try and depict the potential change capacity.

However, what the change heatmap actually depicts is the amount of change impact the various initiatives have added up together.  This shows how much change impact there is and not the actual capacity that the impacted stakeholder groups have.  It could be that certain parts of the organization are agile, mature and have great leaders. Therefore, they are able to have a much greater capacity to undergo larger volumes of change than another part of the organization.  To read more about change heatmaps go to The death of the change heat map.

To resolve this it is important to map out the level of change capacity.  How does one do this?  By using historical data and comparing the level of change against business feedback such as performance indicators, and employee and leader feedback.  To automate this process whereby you’re able to visualize the impacts of change against the plotted change capacity levels of each part of the business leverage The Change Compass.

5. Change prioritisation

Respondents call out the fact that often prioritisation of initiatives is made based on typical project manager data points such as cost, timeline, funding and business results.  The gap is that change data should also be taken into account.  Data such as the velocity of the change, the volume of the change, change capacity, risk of impact on business performance, business readiness, all should be valid data points to consider in making prioritisation decisions.

With the ability to access a range of data points, the organisation is better able to make balanced decisions to maximise benefits and minimise risk.  The fact is that with the various challenges listed here in not being able to access a range of change data, decision-makers simply make decisions based on whatever they can get their hands on. 

6. Data recency and validity

The usefulness of data is only as good as its recency and validity.  Outdated data cannot be used to make decisions.  What respondents call out is that it is difficult to ensure that data is constantly updated and valid.  Once again, keeping data recent takes significant time and effort.  However, various digital tools can again be leveraged to support data recency.  At The Change Compass we build in a feature to remind users to update information and data recency is also depicted in reports to reinforce the update of data.

To read more about change data to go The ultimate guide to measuring change.

7.  Change governance

Change governance is critical to be able to support and govern the change data collected and reported.  Change governance does not need to be a separate body created just for the purpose of governance change data.  It could be a business unit planning meeting or a part of a PMO agenda for example.  The purpose of the governance body here is to reinforce the importance of data, review any generated insights, and make decisions on how to apply insights to business decisions.

To read more about change governance, how to set it up and various roles and responsibilities, refer to our Ultimate guide to change portfolio management.

As a change community, our challenge ahead remains how we adopt and embrace the new world of data and insights.  The more we are able to leverage data and not shy away from it.  The more we are able to move the discipline forward to that which is seen as directly driving business value and has a critical seat at the table in decision making.

Five agile change toolkits

Five agile change toolkits

The agile way of implementing changes has been popular for quite a number of years among a range of companies, from small startups through to large corporations. Most agile methodologies do not address the role of change management explicitly as a function. However, at the same time, most project practitioners agree that managing change is a critical skillset. In fact, surveys conducted by the Project Management Institute consistently found that change management is rated as one of the top skills for a project manager.

To find out more about agile methodology and embedding change management within it, please read our Ultimate Guide to Agile for Change Managers.

In this article, we will focus on a range of toolkits that support agile to help change managers implement change. Gone at the days when the change manager needs to work on large presentations and slides detailing every aspect of the plan. It was not uncommon to see more than 100 slides for a change plan. In the agile world, documentation is important but more important is the conversation and working with stakeholders.

Toolkit 1: Change Canvas

The change canvas or ‘change-on-a-page’ is a summary of the change plan. It follows a similar simple and summarised format as a Lean Canvas. The change canvas may be used to socialise what the change is about and the approach in implementing the change with a range of stakeholders.

Previous versions of the change canvas are often designed with more of a project plan slant. In the current version, we focus on a core set of questions that the change practitioner needs to answer in creating a change plan. To download the canvas click here.

Toolkit 2: Change experiment card

A core part of agile is about experimenting and iterating through a series of changes, versus planning one change. The idea is that each small change is an experiment with a hypothesis that can be tested and proven to be true or false using data. When the overall change becomes a series of smaller changes, each change iterates on the previous change. The overall risk of failure is reduced and each change is one step closer toward the ultimate successful end state.

Applying this concept in change management – The change experiment card is a template to help you design, plan and test your change experiment. To download the template please click here.

Change experiments can range from:

  • Project message positioning to stakeholders
  • Learning design effectiveness
  • Effectiveness of a communications channel in engaging with stakeholders
  • Change readiness tactic
  • Effectiveness of the change vision artefact

Toolkit 3: Behaviour over time graph

Plotting expected or actual stakeholder behaviour over time is an effective way to anticipate or track how they are experiencing change. It can provide significant insight on whether additional change interventions are required to shift the stakeholder towards the change process effectively, if there are any obstacles being faced or if the time taken along the change journey is the speed as anticipated.

Here is an example of a behaviour over time graph.

Change readiness over time plotted against change milestones

Toolkit 4: Connected circles analysis

The connected circles analysis chart can be used to understand the influencing powers of various stakeholders within the project. Agile projects are very much dependent on effective stakeholder engagement and collaboration. A range of stakeholders are thrown together within the same project from the beginning and there is a high expectation of successful collaboration and teamwork across the board. This analysis helps you to visualise the power dynamism and influence mechanisms amongst different stakeholders.

With the insight gained from this, the change manager can better focus on how to resolve any relationship issues, risks, and leverage the network to achieve better relationship and outcomes within the group.

Project stakeholder influence dynamics

Toolkit 5: Causal loop diagram

Systems thinking is critical in agile projects. Systems thinking means that you’re able to see the various components and how these components affect each other within the overall environment, or system. This contrasts with a linear view of A causing B or vice versa.

The causal loop diagram helps to flesh out and analysis key factors in the overall system and what causal relationship there are between different factors.

The below example shows employee sentiments toward a system change. This is a very simplified version of what happens since in real scenarios there could be various factors that are reinforcing each other, leading to lots of arrows pointing at different directions. At a more sophisticated level, you may assign points in terms of the strength of the causal relationship. At a basic level even plotting the causal relationship between a few key factors may generate key insight into the ‘why’ of the dynamics of a situation.

Causal loop diagram of the rationale of stakeholder sentiments

For those who work in organisations that are undergoing a significant number of agile changes, there needs to be a way to capture and visualise these changes so that the data can aid decision making for stakeholders. Using data visualisation, stakeholders can gain a better grasp of the various changes across the organisation and be able to understand key capacity challenges, crunch periods, the velocity of changes across time, and pinpoint particular parts of the organisation that may need extra support.

The following are key steps in which an organisation can leverage tools such as The Change Compass to derive one view of change and to better plan the implementation of changes. With embedded operational routines that regularly focus on change data in conjunction with other business and project data, the business is able to build its change capability through constant reviews, valuable stakeholder discussions, iterations on change tactics and adjusting plans to get ready for change.

To download this diagram click here.