I’ve been working in the Pacific for the last 20 years designing and building information systems. One of the key issues that keeps cropping up is the problem of focusing too much on designing and building systems that are highly efficient. There seems to be a general lack of understanding that focusing too much on efficiency creates problems.
The basic rule of thumb is that generally speaking the more optimised a system becomes, the more brittle it becomes. This is because as a system that is over optimized doesn’t have any redundancy options available when the environment changes.
You see this pattern playout consistently across:
Ecological systems – when apex predators who are over optimized for their environment go extinct due to changes in their environment
Stock market systems – when automatic trading algorithms go haywire due to some unexpected event and have run away feedback loops that results in flash crash scenarios
Software Systems – that are over optimized for a specific use case that become redundant when user requirements change
To successfully implement solutions, it is critical to understand what the most important problems are and what should be tackled first. To do this we will need to create a shared understanding of what the priorities should be via a Pain / Frequency Matrix.
A Pain / Frequency Matrix helps us develop this shared understanding by providing a framework for assessing what the most painful problems are and how often they occur. We want to focus on the most painful processes that occur frequently.
The objective of the exercise is to create a pain / frequency matrix for each solution identified in an empathy map by:
Ranking how painful a problem is.
Ranking how frequently a problem occurs.
Plotting the pain and frequency for a given problem onto a pain and frequency matrix to determine which problems.
3 different colored sticky notes
Empathy Mapping Canvas template (optional)
Miro Software (optional)
Place the green sticky note that has name of the stakeholder in the middle of the matrix.
From the empathy maps take each of the red colored sticky notes (i.e. the pains / issues) and place it in one of four quadrants depending on how painful and how frequent an issue happens:
High Pain / High Frequency: Place the issues that you think are very painful (i.e. they create lots of inefficiency / work or lots of mistakes) and occur often in the High Pain / High Frequency quadrant.
High Pain / Low Frequency: Place the issues that you think are very painful (create lots of inefficiency / work or lots of mistakes) but don’t occur often in the High Pain / Low Frequency quadrant.
Low Pain / High Frequency: Place the issues that you think are not painful but occur often in the Low Pain / High Frequency quadrant.
Low Pain / Low Frequency: Place the issues that you think are not painful but occur often in the Low Pain / High Frequency quadrant.
You should now have a pain and frequency matrix that looks something like the image below:
To figure out how to improve a process, we must first understand what issues/problems/pain points each stakeholder in the process faces. To do this we will need to create a shared understanding of what the issues are for each stakeholder involved in the process. An Empathy Mapping helps us build this shared understanding.
The objective of the exercise is to create an empathy map for each stakeholder in a Process Mapping Canvas by:
Mapping out what task each stakeholder does? Use the Tasks from the Process Mapping Canvas.
Mapping out what pain points a stakeholder faces when trying to do those tasks? Pain points are issues that prevent stakeholders from improving the process KPIs outlined in the Process Mapping Canvas.
Mapping out what the gains would be if we were able to solve the pain points? Gains are the improvements in the process KPIs if the pain points were to be eliminated.
3 different coloured sticky notes
Empathy Mapping Canvas template (optional)
RealtimeBoard Software (optional)
Each stakeholder takes their role (green sticky notes) and places the role at the top of the Empathy Mapping Canvas. They take the tasks (yellow sticky notes) and place those in the tasks area of the Empathy Mapping Canvas.
For each task, stakeholders identify issues they have with performing the tasks and add these issues to the pains area of the canvas. Issues raised should have some sort of negative impact on a processes KPIs. Ensure that there is one issue per sticky note.
For each pain, stakeholders identify what the gains would be if we were to solve their pains. Gains identified should have some sort of positive impact on a processes KPIs. Ensure that there is one gain per sticky note.
When completed you should have something like the above. This is the empathy mapping canvas.
To continually get better at what we do, we need to be able to map out what it is we are doing, who is doing it and how we are measuring its performance. A Process Mapping Canvas helps us understand who is doing what in a specific process and what the performance metrics are in terms of efficiency and effectiveness.
The objective of the exercise is to:
Map out who the stakeholders are in a process (Green Sticky Notes)
What tasks do they do (Yellow Sticky Notes)
Determine what the metrics are for efficiency and effectiveness (Blue Sticky Notes)
3 different coloured sticky notes
Process Mapping Canvas template (optional)
RealtimeBoard Software (optional)
Fill in the details of your process. In the example below we are mapping out a travel requisition process. We’ve set the date of when the process was mapped out, when it will next be reviewed, the divisions that this process applies to and who was involved in the mapping.
Take the green sticky notes and write down the roles of every stakeholder that is involved in a process. The rule is to have one sticky note per role.
Take the yellow sticky notes and write down what tasks each stakeholder does. Place these tasks directly below each stakeholder.
Take the blue sticky notes and write down what the efficiency and effectiveness metrics should be for the process.
The formula for efficiency is total output / total input (usually time or effort). E.g. If your process is outputting Permits then your efficiency metric could be total # of processed permits / total time taken to process permits. For more information on the theoretical framework for this you can look into efficiency formula
The formula for effectiveness is total desirable (or undesirable) output / total output. E.g. If your process is outputting Permits then your effectiveness metric would be total # of approved permits without violations / total # of approved permits. For more information on the theoretical framework for this you can look into confusion matrix
When completed you should have something like the image below. This is the process mapping canvas for a travel requisition process (i.e. staff requesting to travel).
The following are additional activities you can do to extend the process mapping exercise further.
Number each task in order of execution
Indicate with blue sticky dots which tasks are done electronically (i.e. with a computerised system)
IT Strategy frameworks are tools that help structure IT practitioners think through actions and investments into information technologies to accomplish specific objectives. They are used to analyze problems and develop strategies, tactics and specific actions to take to resolving these issues. The following frameworks are the key frameworks that we use to think through from a strategic level how we approach IT strategy.
IT strategic thinking, decisions and implementation methods in the Pacific are typically based on very poor theoretical foundations. This leads to high risks of failures in programmes and projects.
Establish a set of common strategic management frameworks to help guide high level strategic thinking across disciplines
The following 5 frameworks outlined below are used to analyse and think through strategic decisions and actions.
What does it do?
Helps us understand how decisions are made and helps us diagnose where things may have gone wrong when problems arise
What does it say?
All decisions cycle through 4 separate and distinct phases. These are: Observation of what is happening in the environment; Orientation and understanding of what has been observed; Deciding what to do based on this understanding; Acting on the decision; These actions then have some impact on the situation and this is then re-observed and the cycle then repeats.
Helps us understand what type of systems we are dealing with and which management methods would be appropriate to apply to each system
What does it say?
There are 3 types of systems, Ordered systems, Complex systems and Chaotic systems. Each one needs to be managed differently. Ordered Systems can be further split into Obvious / Common Systems and Complicated systems.
Helps us map and understand what an operating environment looks like and where we should be investing our resources
What does it say?
All things evolve through 4 distinct phases driven by supply and demand competition. For technology these 4 stages are: Stage 1. Scientific Research is conducted into a some phenomena to understand it, Stage 2. Using the understanding of this phenomenon, tools are created to harness this phenomenon to solve some real world problem Stage 3. If valuable (demand side), more and more of these custom built solutions are created (supply side) and a market begins to from. As the market develops competition begins to intensify and then stabilise around a common design and the solution becomes a mass market product with very little differentiation. Stage 4. Eventually the solution becomes so standardised and commodised that it becomes a utility service, taken for granted by the general population until it is no longer available.
Helps us understand if a particular solution should be implemented
What does it say?
All system designs MUST satisfy 3 conditions before implementation can begin Desirable – The solution solves enough of a problem for people that they are happy to allocate resources to its development and maintenance Feasible – the solution can be implemented with the allocated resources Viable – the solution can be sustained long term because the revenue streams exceed the cost structure of the system
Helps us understand and communicate the scope of a government system as well as the order in which it’s constituent components need to be implement and why they need to be implemented in a particular order
What does it say?
All governance systems have 3-axis in which scope can be understood. These axes are the Governance Domain, these are the places where constraints can be imposed on a system to influence the behaviour of the entities operating within it; Governance Functions, these are the tasks that are carried out by governments to ensure the entities adhere to the rules; Governed Entities, these are the entities whose behaviour we are trying to influence
Helps us understand how to architect a data warehouse system.
What does it say?
Data marts are repositories of data belonging to particular lines of business. The data warehouse is simply a combination of different data marts that facilitates reporting and analysis. Based on Ralph Kimballs “bottom up” approach (Note that this is in contrast to Bill Inmons “top down” approach, (i.e. the Data warehouse is the centralised repository, data marts are created from it)
With all the hype surrounding big data it’s difficult to consider what other alternative approaches there may be to gain insight into what is happening within a society. In government the traditional approach of data collection focuses on survey data or data generated from government information management systems that have been aggregated into a centralised data warehouse. These are then analysed and used for informing policy. The problem is people lie on surveys all the time and operational information management systems only capture a small part of what’s really going on.
Anecdotes and stories are rich in insights. Unfortunately collecting anecdotes using traditional methods doesn’t scale well and people’s biases tend to get in the way during the collection and analysis process. If we could scale the capturing of anecdotes and make sense of it at scale as well as develop a methodology to account for bias it would greatly enhance the sense making ability of policy makers. That’s what I’ll outline below and show you a demonstration of an app to do just that.
Government information systems in the Pacific are too often ‘built to last’ rather then ‘built for change’, which is the real requirement for them to be.
The traditional approach has been to build very tight vertically integrated systems. This has created unintended ripple effects that cascade across the system whenever a change is made. So what typically should have been a simple update of a business policy, process change or business rules requiring a few hours’ work often times turns into a complex, bureaucratic and code-intensive process that instead takes months of tedious effort to implement.
A core problem we find with e-Government initiatives in the Pacific is the lack of clearly articulated business models. Well thought out business models ensure that any e-Government or digital transformation initiatives remain sustainable once the initial project funding runs out. Unfortunately most government departments in the Pacific don’t have a framework in place to develop business models for their e-Government or digital transformation projects.
So we’ve adapted the business model canvas framework from Strategyzer to help our clients understand and then develop business models that are sustainable for their e-Government projects. For those of you not familiar with the business model canvas from strategyzer take a look at their introductory video.
We’ve taken this initial business modelling framework and then expanded on this to ensure that it fits into the e-Government context. We will release a series of blog posts later that expand on how we’ve modified the business model canvas.
We help you conduct a situational analysis of the current operating environment based on the needs of the system you’re trying to implement. This involves assessing the current technological landscape and climate and then mapping out the value chains that will be required to create the system. To do this we use a technique called wardley mapping.
Wardley mapping was created by researcher Simon Wardley who is an evolutionary biologist by training. Here he explains the technique at OSCON 2014.
Here’s a simple visual diagram that we use with clients to develop their strategies.
When helping you develop a strategy, it is absolutely critical that we build up a shared understanding among stakeholders of the following components of a strategy:
This is how we facilitate the creation of your strategy. We create design thinking workshops to
Process Mapping: Map out your current processes, whos involved and what is it that they do
Performance Mapping: Then we define what the processes performance metrics are for effectiveness and efficiency.
Empathy Mapping: We then create an empathy map for each stakeholder involved in the process to get an understanding of what issues they are facing.
Value Chain Mapping: From here we can identify the components needed to create a solution to address the issues identified for each stakeholder.
To effectively build a shared understanding of your strategy we follow a few key design principles:
Group Workshops rather than individual desktop work
Face to Face interactions rather than emails
Filling out Visual Frameworks rather than Report Writing
We’ve listed below a few concepts that we will need to establish in order with you to ensure we are all on the same page.
When working in government, you are typically in the business of shaping the behaviour of the entities (citizens, companies etc.) your department is responsible for.
Your goal typically is to create rules that ensure that amplify good behaviour and dampen bad behaviour within the jurisdiction you are responsible for. Before developing strategy you need understand and be aware of how rules affect behaviour. This is difficult to do as the consequences of rule changes may have unintended negative consequences. We’ll show you how to do this using by running multiple safe to fail experiments in sandboxed environment.
To develop strategy you must first understand the landscape you are operating in. To do this you must:
Know your users (who are you serving) – covered by process mapping
Know their needs – covered by empathy mapping
Know the prerequisite activities to meet those needs – covered by value chain mapping
Add position (connect users, needs, and prerequisites from top to bottom according to dependence) – covered by value chain mapping
Add movement (place needs and prerequisites left to right according to evolutionary stage) – covered by value chain mapping
In government the environment in which you operate typically consists of:
Governance Stack; the mechanisms used to shape behaviour e.g. legislation
Governance Functions; the activities the government carries out in order to shape behaviour e.g. Registration, Authorisation, Monitoring, Inspection, Prosecution etc.
Governed Entities; the actors whose behaviours we are trying to shape e.g. Citizens, Companies etc.
Governing Entities; the guys who decide what the rules should be e.g. Political parties / Politicians / Policy Makers, Regulators, Government Departments etc.
The Governance Stack can be broken down into:
International and Regional Agreements
Legislation and Regulations
Processes and Procedures
The Governance Function can be broken down into:
The Governed Entities (and Assets) can be broken down into:
Natural Persons i.e. Citizens
Legal Persons e.g. Corporations, Companies, Organisations etc.
Assets e.g. Land, Property, Vehicles, Vessels etc.
All things evolve and this is true in technology. Technology evolves in a predictable fashion.
Technology typically starts with scientific investigation or research into a phenomenon. As scientists build up an understanding of the phenomenon further novel uses are discovered and developed for application of the phenomenon to problems. These novel uses are the birth of the technology (research phase).
Based on this research an understanding begins to grow on how to apply the technology to various other problems. A few scientists, engineers and hobbyists begin to use this understanding to create custom built applications for the use of the technology (development phase).
As development accelerates a market begins to form around the technology, a few companies emerge to provide the technology to the masses as a commodity product. Over several decades the market around the technology drives it the point where it is so commodised (cheap, reliable, repeatable) that it is used as a building block for other technologies. (market phase)
Some technologies evolve to such a highly commoditized stage where they can be delivered as a utility service (utility phase).
The table below contains a list of characteristics to help you determine how evolved something is.
Stage of Evolution
Slowly increasing consumption
Rapidly increasing consumption
Widespread and stabilising
Rapid increases in learning
Rapid increases in use / fit for purpose
Normally describe the wonder of the thing
Build / construct / awareness and learning
Maintenance / operations / installation / features
Focused on use
Learning on use
Learning on operation
Known / accepted
Domain of experts
Increasing expectations of use
Ordered (appearance of being linear) / trivial
Different / confusing / exciting / surprising
Leading edge / emerging
Common / disappointed if not used or available
Standard / expected
Perception in industry
Competitive advantage / unpredictable / unknown
Competitive advantage / ROI / case examples
Advantage through implementation / features
Cost of doing business / accepted
Focus of value
High future worth
Seeking profit / ROI?
High volume / reducing margin
Poorly understood / unpredictable
Increasing understanding / development of measures
Increasing education / constant refinement of needs / measures
Believed to be well defined / stable / measurable
Constantly changing / a differential / unstable
Learning from others / testing the water / some evidential support
Essential / operational advantage
High / tolerated / assumed
Moderate / unsurprising but disappointed
Not tolerated, focus on constant improvement
Operational efficiency and surprised by failure
Gambling / driven by gut
Exploring a “found” value
Market analysis / listening to customers
Metric driven / build what is needed
Reducing the cost of change (experimentation)
Reducing cost of waste (Learning)
Reducing cost of waste (Learning)
Reducing cost of deviation (Volume)
Heritage / culture
Analysis & synthesis
Analysis & synthesis
The movement of a component along the X axis is determined by its stage of evolution.
The following are patterns that are very important to learn.
Pattern 1: All things evolve
Everything evolves from left to right under the influence of supply and demand competition.
Genesis Unique, rare, uncertain, constantly changing, newly-discovered.The focus is on exploring.
Custom Uncommon, frequently-changing, requires artisanal skill, no two are the same.The focus is on learning and developing the craft
Product (and rental) Increasingly common, more defined, better understood. Repeatable processes. Change is slower. Initial differentiation but increasing stability and sameness. There are often many of the same kind of product.The focus is on refining and improving.
Commodity (and utility)Scale and volume operations of production. Highly standardized. Defined. Fixed. Undifferentiated. Fit for a specific known purpose. Repetition, repetition, repetition… With time, it becomes commonplace and less visible.The focus is on ruthlessly removing deviation, industrialising, and increasing operational efficiency.
Pattern 2: Characteristics Change
As components evolve, their characteristics change.
Cost of Doing Business
The training of your people, the standard ways of operating, and the techniques that you almost always apply. Select cells multiple times to progress through colors indicating a weak, warning, good, and neutral (undetermined) status.
Phase 1: Stop Self Harm
Use a common language (necessary for collaboration)
Challenge assumptions (speak up and question)
Focus on high situational awareness (understand what is being considered)
Know your users (e.g. customers, shareholders, regulators, staff)
Focus on user needs
Remove bias and duplication
Use appropriate methods (e.g. agile vs lean vs six sigma)
Use a systematic mechanism of learning (a bias towards data)
Think small (as in know the details)
Phase 2: Becoming More Context Aware
Be transparent (a bias towards open)
Focus on the outcome not a contract (e.g. worth based development)
Be pragmatic (it doesn’t matter if the cat is black or white so long as it catches mice)
Use appropriate tools (e.g. mapping, financial models)
Think fast, inexpensive, restrained, and elegant (FIRE, formerly FIST)
Use standards where appropriate
Move fast (an imperfect plan executed today is better than a perfect plan executed tomorrow)
Strategy is iterative not linear (fast reactive cycles)
A bias towards action (learn by playing the game)
Manage inertia (e.g. existing practices, political capital, previous investment)
Effectiveness over efficiency
Think aptitude and attitude
Think small (as in teams, “two pizza”)
Distribute power and decision making
Phase 3: Better for Less
Be the owner (take responsibility)
Think big (inspire others, provide direction)
Strategy is complex (there will be uncertainty)
Commit to the direction, be adaptive along the path (crossing the river by feeling the stones)
Be humble (listen, be selfless, have fortitude)
A bias towards the new (be curious, take appropriate risks)
Optimise flow (remove bottlenecks)
Do better with less (continual improvement)
Set exceptional standards (great is just not good enough)
Seek the best
Provide purpose, mastery, & autonomy
Phase 4: Continuously Evolving
Exploit the landscape
There is no core (everything is transient)
Listen to your ecosystems (acts as future sensing engines)
Design for constant evolution
There is no one culture (e.g. pioneers, settlers and town planners)
By examining the doctrine in an organization, you can get an idea of how adaptable it is and how well it will respond to external change or gameplay. You can do this with your own organization, or with other organizations.
In-person? Gather several people from different levels of the organization and perform the above self-assessment together. (There may be arguments, but that’s not a bad thing.) Distributed? See this form-based assessment by Justin Stach.
Once you’ve assessed the status quo of doctrine in your organization, you can go about addressing areas of weakness. Simon suggests you do this in phases. The above self-assessment’s phases presents his best guess at the order in which you should tackle them.
The context-specific strategy you choose after considering your purpose, the landscape, the climate, and your capabilities.
As an example, where might you focus in the below scenario (view in MapScript) if you wanted to increase competition around content?
To increase competition in content, the obvious option is to cause the industrialization (evolution) of the creative studios. More competition among studios would result in more, better content, so how could we make that happen?
A naive move might be to launch an independent creative studio or form a strategic partnership to advance one particular studio, but that game can all-too-easily be lost. There are more interesting options in the lower-level constraints (a Fool’s Mate).
If there were more competition among production systems, for example, the barrier to entry for new talent and therefore new creative studios would be lowered. An open approach would accelerate that process, indirectly causing increased competition among creative studios and ultimately content.
Chances are, the existing creative studios won’t have the situational awareness to recognize the play for what it is. In fact, they might support it in the name of short-term cost savings. Wild! Read more on this scenario in Simon’s post, Fool’s mate in Business.
Our purpose and the landscape
The context and how it is changing
Our level of understanding of the environment
The map in use
Uncharted vs Transitional vs Industrialised
Of evolution e.g. Genesis, Custom, Product, Commodity
Activity, Practice, Data or Knowledge
A single entity in a map
The user need
Position of a component relative to the anchor in a chain of needs
Something a higher level system requires
High level needs you provide to others
How evolved a component is
Connection between components
Transfer of money, risk & information between components
Rules of the game, patterns that are applied across contexts
Approaches which can be applied regardless of context
In this article we discuss how to use the business model canvas to work out whether the system we are implementing makes sense in a government setting. Traditionally the problem with the the Feasibility part of the business model canvas is that it doesn’t work for the typical government department. Note that the Business Model Canvas can be split into 3 major sections:
Desirability – does the system deliver enough value that stakeholders will pay for it?
Feasibility – can the system be built with the resources available?
Viability – once built can the system be sustained from the value its created for stakeholders?