3. The DADA Loop — How To Improve Decision Making In Organisations
As the pace of change continues to intensify, there is relentless pressure to improve delivery effectiveness. Budget cycles bake in assumptions of performance uplifts, volume growth, operational savings or all of the above. This often results in a mass of KPI’s that may in isolation be well-intentioned, but can be overwhelming when they combine at the delivery level.
The problem with many of these targets is that they are applied from a position far removed from the reality of the teams having to ensure their implementation. This usually gives rise to a process optimised for ease of reporting by those central teams, which is problematic when these reports become the only management KPI’s steering the organisation. The result is often turning what should be a continuous improvement mindset into a mass of separate KPI mini-projects chased up by different managers competing for attention.
This area is the remit of the Management Information System (MIS), an area surprisingly often overlooked due to a focus other reporting requirements — accounting, environmental, HR, etc. These reports are clearly essential to meet legal obligations, but they serve a different need than those required for a comprehensive MIS. Accounting cycles for example are slow and often produce numbers that have been manipulated to meet standardised financial requirements making them of little use to the management of continuous improvement processes.
To be effective MIS must alert management when a problem has occurred. To do this, systems should use actual performance to distinguish between activity that is productive, non-productive, or when no activity has occurred. An effective MIS will provide a continuous monitoring through the complete cycle from forecasting, planning and delivery, allowing rapid feedback and intervention. By providing this continuous cycle, teams are able to see whether the decisions they have made have been effective.
Differences between fixed outcome and continuous improvement systems
Fixed outcome measures are common as they present a simple model and allow for basic reporting. These reports assume that a fixed and pre-determined solution will deliver the needs of the business. A valid example would be an audit closure tracker, where once a defined action has been validated, the point is closed. The danger is fixed measures are often reductionist in their assumptions, acting to focus the organisation on task completion rather than provable outcomes. For example, if an audit point action does not actually remediate the issue, this can trigger a conflict over closure of the KPI (wasteful infighting) rather than using this as a learning opportunity for the organisation (risk reduction).
Continuous improvement approaches instead rely on a robust feedback mechanism. This allows the testing of ideas and a change in approach when needed. These systems do not assume the selecting the perfect next step, but instead on rapid feedback to determine success and speed up the learning process. This is the basis of truly effective decision making, and is where the Data-Analysis-Decision-Action (DADA) loop comes in.
Data — Analysis — Decision — Action: The DADA loop
This loop is needed at all levels of the management structure to provide the timely feedback and self-correction mechanism for decision making. The approach is equally valid whether applied to external facing functions e.g. forecast vs. actual sales, or internal processes e.g. required vs. actual delivery performance. The DADA loop is deceptively simple as a concept, but making it function effectively is one of the distinguishing features of world class organisations.
Data
At the core of the feedback loop is the need for decisions to be validated by data — a process divorced from data is not a robust decision-making process.
Imagine being a passenger in an aircraft where the captain strolled around making some of the following statements. “I think we have enough fuel to make it to our destination” or “I always assume the weather is going to fine”. You should feel nervous! You would probably feel that many of the decisions concerning your safety were being made on unreliable data. Similar situations occur in companies concerning tough decisions or trade-offs. To be competitive today means managing the business on readily available facts — not perceptions based upon “experience”.
Many companies actually suffer from too much data. Deciding which data to collect is not judged by volume or ease of availability — rather it has to be judged by the quality requirement and needs of the organisation. Relying on junior team members whose main skill may be proficiancy with a data visualisation tool is foolish — the changes needed to allow rapid feedback can be impacted by in governance and control processes, management structures and regulations. This is core responsibility of the mmagement team to address and implkement.
Any operation can be regarded as having two major focuses. There is a process itself and there are the people who support the process. Data is required both on the process element because this reflects the result of the activities of the supporting function, and of the people element because this indicates how and why the process has performed as it did.
The amount of work planned will always be a key piece of data. This is the work that was considered to be necessary in order that the process could have performed at optimal efficiency. The amount of actual work achieved is of interest because this will indicate whether the plan was not achieved, met or exceeded (see previous posts on overcommitment for the risks this creates).
The demand and the availability of skills will be another key data point, as the demands on the workforce inevitably evolve over time. This data should be tied to hiring, training and personnel improvement plans.
Analysis
Data collected around the parameters previously mentioned will not be meaningful to management and supervision. They need to see trends and variances and it is here that effective systems carry out a vital function. Effective systems translate data, via analysis, into trended Key Performance Indicators
The stage of analysis takes raw data and transforms it into information so that management may quickly and easily identify both trends and variances from the desired performance. The analysis process in the DADA circle creates clarity from a mass of data.
For example, plan attainment helps improve the forecasting process by comparing it to what was actually achieved. Quality measures are an obvious choice if we wish to attain a “get it right first time” culture. In production systems, Mean Time to Repair would give an indication of whether planned times were being achieved and would also indicate what influence maintenance might have upon the production systems availability.
Decision
Given a firm base of data and an effective analytical process, decision making becomes more predictable. Management is able to address quantified problems, and the predictability of possible improvement can then be forecast accurately. This does however need enough psychological safety for people to act on the analysis no matter where that may lead. For example, it can be tough to abandon a product that is not delivering, or a previously selected solution that is not meeting requirements, if this choice will leave the team feeling exposed.
The system will support management in that it will identify the source of the problem. For example, a robust system will identify that a certain proportion of capacity is being lost due to an activity. This does not take away the need for management and technical expertise necessary to identify why that is the case. It allows management and technical expertise to be focused more effectively, and to make a timely decision.
Action
An area which constantly identifies world class performers is the ability of the organisation to drive home decisions by turning them into actions. World class performance means having the ability to make things happen.
Although the process is not complex many companies fail to master it. What needs to be done — who will do it and when will the actions be completed? The secret of success is to break the action down into detail and to have systematic reviews that takes place frequently. In this way any off-schedule conditions can be corrected before too much slippage has occurred.
Clarity of communication and objectives is also a critical part of action delivery. What is the expected change we hope to see, and how will this be reflected in the data? This is where a rapid feedback loop becomes essential — as new data becomes available it allows the action effectiveness to be understood
In Summary
The circle is complete. The systems are designed to collect the data that will be required to make robust decisions which is then converted into trends and variances that help identify and quantify problems that occur. The system, if effectively designed, will aid the decision process — but the real secret of success is turning those decisions into concrete action.
