Model for Improvement

One of the most common components of political or professional promises is improvement. This is either the process or the fact of getting things better. An equally common component of the subsequent period after election or appointment, is the lack of improvement. Why is this?

It is easy to blame lack of interest, defective structures, lack of support, or simply incompetence as the reason. However, the issue may well be not knowing how to bring about improvement in a sustainable way. The Associates in Process Improvement (API) developed a a model to address this very problem, and its quite simple. This model is named, “Model for Improvement.”

Underpinning any improvement plan are a couple of points. First and foremost is the saying that if you can’t measure it, you can’t improve it. This is not entirely true, but what is true is that if you can’t measure it, you cannot show or prove that you have improved, nor are you likely to know whether you have improved or not.

Questions underpinning improvement

Secondly, there are four questions that improvement processes require to answer:

  1. What is the problem that we are trying to solve?
  2. What are we trying to achieve?
  3. What steps or actions can we take to realize the achievement?
  4. How will we know when we get there?

You ought to answer these questions before you effect any changes otherwise, you run the risk of embarking on a wild goose chase and wasting the resources of you employer or country. Collectively, they constitute the first part of the model for improvement. The second part is what is called the Plan Do Study Act cycle (Also called Plan Do Check Cycles)

PDSA cycles

The PDSA cycles are a repetetive four stage model for implementing an improvement. Once you know what you want to achieve as required by the questions listed above, you make a plan.

Your plan should include SMART objectives in line with your overall goals, strategies, actions, a projection of ALL the resources that will be required to successfully implement the plan, the time required, the people responsible for individual tasks in the plan, the potential risks the plan may phase at implementation, and the actions you project to manage those risks. The plan also includes an endpoint, and measurement tools to objectively assess the impact of the various actions/activities set out in the plan.

Once the plan meets your satisfaction, you proceed to the “Do” phase. The do phase is where you implement the actions and activities set out in your plan. Make sure to have a baseline documented. This should be obtained at the planning phase, but if that is not done, it ought to be the first part of the do phase. The baseline tells you where you were before any of your interventions was implemented as is a necessary reference point.

Once the activities are implemented, you proceed to the “Study/Check” phase. Is the projected impact being witnessed? What was our baseline when we started? Where are we now? Is it an improvement? The result of the study phase will inform the next phase. During Study, you may find that the intervention results in a worse situation, no change, modest improvement, or a marked improvement. A key component of the study phase is the root cause analysis. When a result is not as expected or projected, a root cause analysis is carried out to understand what factors resulted in the deviation from the expected. The idea is to modify the root cause to have better outcomes. This ushers us to the Act phase. In the event of deficiencies being observed, remedies as can be made based on the findings of the RCA. Bear in mind that acceptable actions may include rolling-back the changes and perhaps doing away with the project altogether. This however requires courage and humility, neither of which is easy to find.

There is no limit to the number of times PDSA cycles that can be implement. Indeed this should be the routine baseline.


An improved version of PDSA is the OPDSA. “O” stands for observation: observe the prevailing situation before you make plan. That sounds good, right?


An separate improvement model is DMAIC from six sigma – Define, Measure, Analyze, Improve, Control. There are important differences in the two processes but they are not mutually exclusive. Indeed all the steps in the two models are useful.

Model for improvement components

What are the critical components required to implement the model for improvement?

  1. The political will to improve and allocate resources towards improvement
  2. Problems that can be solved by improving processes in a measurable way.
  3. People to carry out the interventions designed in the PDSA cycles
  4. Pragmatic interventions that can be implemented within the resources allocated to the improvement process
  5. Measurement tools that can be used to determine the baseline as well as assess progress.
  6. Time. Remember to set timelines to all improvement processes at the time of planning.


The model for improvement can be implement in a wide range of scenarios: from improving your own daily habits or routines, to improving productivity in a small scale business, to improving quality and outcomes in healthcare to achieving one of a country’s big four agenda. The process is simple but rarely applied, perhaps because the model is not well known across board.


A few years ago, to address the matter of congestion in Nairobi, the ministry of Transport banned 14 seater matatus and replaced them with bigger capacity buses: 25 seater and above. Did this intervention follow the model for improvement? How would you apply the model for improvement to this issue if you were the one in charge of policy? Leave your response in the comments section and Look out for my blog post on this very matter for us to compare notes.

Please follow and like us:

Leave a Reply