|
There is no single and standard scorecard generation process. A lot depends upon the automation and analytical support you have. This page provides the complexities involved in the process, the components of a scorecard, process of generation, notes on scorecard and communication dynamics during the process. Please do refer designing scorecards before proceeding on this page.
Scorecard generation is complex & difficult to automate due to following reasons:
Please refer our field tip Scorecards need manual finish
Audience Specific formats:
A CEO OR a business head has his own very specific likings in the way the scorecards will look like. Scorecard will need to align to that format, for the sake of acceptance and involvement.
Non-production and non BI based data sources
A scorecard includes data like:
- Bench-marks, standards
- Non-production data like training data, leads data, sales pipeline data etc...
- Data related business projection for the future periods
Data sets like above are mostly outside of your BI environment. You typically put this data in your performance reporting tool. The performance reporting tool combines the BI based data with this non-BI based data to generate initial sets of scorecard objects. However, to prepare this non-production data to enter into your performance reporting tool, one has to do manual work. For example- let us say that you have 50000 training data records on 10000 sales staff and 3000 new employees in previous quarter. To prepare the data for your performance reporting you may have, to create a summary data set of 500 records (say). Consolidation will also need you to cleanse the data etc. Leaving aside the scorecard requirements, this data preparation exercise has a merit in itself.
Interpretation and analysis
Unlike a dashboard, a scorecard has a major proportion of its content on analysis and interpretation of the performance (refer Scorecards vs. Dashboards). This includes answers to the following questions:
- What is happening? - Commentary on the performance, and how does on view the variations from expectations.
- Why is it happening? -Â Analysis on the reasons, key low and high areas which are contributing to under OR over performance.
- What are we going to do about it? - Analysis and description of the solution (if not solutions, at least the next steps OR discussion point).
- What help and support you need for your actions? - Description of help items and support one needs from the manager OR the peers.
- What is projected performance? - The projections on how the KPIs will move over coming months and quarters. This projection keeps on getting revised in every periodic review.
- What is the status of the actions we agreed in the previous review cycles?
Above components, includes an effort of analysis, articulation and also validation. As you create your commentary you want it to be reviewed by your peer stakeholders, and also by your manager. This is achieved either by a work-flow OR by other collaboration techniques.
How your scorecard will look like?
To keep it very simple, your scorecard will typically have the following areas:
- The data presentation area
- 'What' of Current state assessment
- 'Why' of Current state assessment
- Key new Actions
- Status of the actions agreed in the previous review cycles
- the future State Assessment
- Help items
- Appendix (details supporting the scorecard)
Process of Scorecard Generation
Raw data preparation for actual vs. standards
A scorecard has got a host of diverse data, and this data comes from various sources. The first step is to prepare the 'actuals' data juxtapose it against the targets and calculate the variance. If you have written procedures and trained manpower, you can generate this raw data (with OR without the automated tools). The raw data will be aggregated data in tabular form.
An end-to-end BI environment will ease good amount of this work. It will feed the data directly into the performance reporting system. However, there are more organization preparing their scorecards using excel, than the ones which have an automated platform.
TIP- If you don't have a BI environment and preparing your raw data using excel, keep a good trail of these files, as it may be referred to during review. Once you have generated the scorecard, you should not let these files to be changed. Keep your 'track-changes' on in these files.
TIP- You should try to get your raw data straight from production reports, instead of creating it from scratch. Sometimes the scorecard preparation teams are not even aware that data needed by them is available in one of the enterprise reports. This approach is useful because the reports would have gone through a good deal of testing and quality checks. If some of the data is not available in the enterprise reports, create additional reports. Another reason is that reports are run on powerful enterprise reporting servers, which generally deliver data to you early.
TIP- There are many mid-points between an excel sheet and a sophisticated performance reporting environment. One such middle point is to use light-weight database engine (as you are not looking for production system robustness), and use excel as the front-end. This enables you to keep a better track and also do a better level of sharing.
Draft Data Presentation, to highlight the exception points
One creates first draft of the raw data in terms of graphics, conditional formatting (red, green, amber) etc. This will allow you to decide on the highlights OR 'exception points' of the performance (where the performance is above OR below the expectations).This step does not have to be done in absolute sequential manner. One can start working on a piece of raw data (as prepared in the first step), as it gets done.
NOTE- Many of the exceptions, as they are highlighted in your formats will not be a surprise. For the performance parameters, which you are tracking on more frequent basis (say daily OR weekly), you would have a good idea about the monthly trend and you would have done some level of analysis already.
TIP- Don't spend too much time in the initial OR work-in-progress formatting (in case you are making it manually), as the final formatting will take its own sweet time. The key here is not to arrive at the final look & feel, but to make it good enough for you to identify the exceptions and highlights of the performance.
'What' of Current State
After you have identified the exception points, one first needs to have a holistic assessment of the current state. This assessment will take note of all the charts, graphs and tabular data to give a crisp summary of the current state. We have seen many scorecards, which have got a dozen dials, line graphs and other visualizations. Though 'a picture says a thousand words' a head of business does not want thousand words. She will look for few key messages on the current state, which are coming out of all the glitzy presentations.
TIP- Articulate 'what of current state', and don't add the 'why' in this section. The idea is for the audience to absorb what is happening.
TIP- 'What' of current state? needs to include impact analysis. So if you say that sales was 5% down, also include the impact in sales numbers, and how much we have to do more next month to cover-up etc.
'Why' of Current State
One is able to start the in-depth analysis of why there was over OR under performance. The analysis has to answer generally the following questions:
- The factors which contributed to the exception points (over OR under performance) in the performance.
- Which factors were out of control, under influence OR with-in your control?
- Key issues OR risks which have resulted in the under performance OR could result in the under performance in the future.
- Key opportunities OR upsides which have resulted OR could result in better than expected performance.
Following the typical cause analysis principle, you need to ask 'why' many a times to reach the root-cause.
- An inadequate cause analysis- Sales are 5% down because we hired 10% below target.
- An adequate cause analysis- Sales are down 5% WHY- because we hired 10% below target. WHY- because it is difficult to hire. WHY- Because our compensation scales are below the new competitors and we are not able to attract good talent. WHY- Because our current financial model is not able afford a higher level of sales compensation.
The objective of this stage is that for each issue, one is able to provide all the root-causes. This stage in reality cannot be separated from the 'key actions' stage, as they are iterative.
TIP- Do a time and effort measurement on your scorecard generation. In an ideal world- raw data generation, and 'what' of current state should not take more than 25% of the overall effort. The subsequent stages of 'why?', 'actions?', 'future projections' and 'help items' should consume most of your time.
Key Actions
This stage is the solution-finding stage so that you are able to address the issue OR a risk, and achieve your goals. The root-case analysis and problem solving is a large subject in-itself. There are scores of tools and methods to approach and resolve a problem. We will not cover the process of analysis and problem solving in this page.
The actions and solution which you will evolve can fall in the following categories:
- To fix the under performance. For example- We are going to engage two more 3rd party recruitment agencies to accelerate our sales staff hiring.
- To leverage better than expected performance. For example- Given the positive demand patters, we are deploying additional sales force in the given location to make the best of the situation.
- To share other new management actions to maintain OR exceed the performance. For example- We are going to create a special group to pilot short listed ideas, so that it does not disrupt the existing business and also create show-cases for good ideas.
You may also like to add:
- The status of these actions like 'started', 'completed', 'yet to start'....
- The expected timeline to complete the action.
TIP- Share the actions which you will 'start doing', 'stop doing' OR 'do differently'. Do not include the actions which you will 'continue doing'.
Help Items and Management Support
Every action needs some level of help and support from the management OR other functions and teams. We assume that in any scorecard, there will be top 5-6 actions which will be highlighted. For those actions, one needs to engage with the peers and understand their level of commitment to support. If you feel that you are not getting the commitment, you can either drop the proposed action OR find an alternative. Another approach is to keep the action in the scorecard and raise the help items in the scorecard review session.
Future Projection of Performance
The core part of the scorecard is to view on what is expected in the future and the level of confidence and risks associated with that projection. Projecting the future performance also disciplines the presenter to be realistic and diligent on his statements. Depending upon the issue and the context, the future projection could fall into following categories:
- A firm commitment
- An expected projection with a defined level of confidence- We have 90% confidence to achieve this sales target at the end of 3 months from now.
- A desired target, but no commitments- We will have an 'internal target' to exceed our sales plan by 15%
Performance projections is a way is to confirm on if your business plan is on track to be met. Your business plan is ideally broken down into month by month. This part of the scorecard can include confirmation OR deviation from the business plan and the reasons thereto. Apart from business plan parameters, there could be other underlying parameters, for which you can give the future projections. When we say business plan, it means the goals for the enterprise, functions, processes and teams.
Scorecard Generation Work-flow and preparation dynamics
While we have shared the sequence of scorecard generation, the entire process is highly iterative. You can expect a lot of interaction and engagements across all the stakeholders to the scorecard. In many organizations, the interactions are triggered from the disagreement on the data itself. For example head of sales may not agree with the order fulfillment TAT figures coming from Ops scorecard. Similarly CIO may not agree with the IT delivery TAT for an Ops project, as mentioned in Ops scorecard.
NOTE- Apart from execution management, scorecard generation also drives the fundamental improvements in our processes and information management. For example- Disagreements on the data, and current state, leads to better definition of the reports and a more effective interaction and engagement principles.
The contents of a score-card, may be prepared and certified by different entities. Having some kind of work-flow mechanism will help. Some of the modern performance management tools have this feature.
So-What Principle
TIP- One way to look at a scorecard is to go with 'so-what?' principle. On every component of the scorecard, one can ask a question 'so-what?' which will work on generating a holistic scorecards. This is an example how it will flow
A well presented actuals data ===so what does it mean? ==> The actuals data compared with the expectations and targets===this is mathematics. So what messages we want to give? ==> The current state assessment ===so what do we do of this current state? ===> Key Actions ===these are actions. So what results these actions will lead to? ===> The future state assessment.
Vertical and Horizontal of Scorecard Generation
An ideal sequence of generating an enterprise level scorecard is Bottom-up. In a way, the lower level scorecard rolls-up into higher level scorecard after a detailed review at a given level. Following are the examples of vertical levels which may exist:
- Team Level
- Departmental level
- Functional Level
- Enterprise Level
The scorecards can also span across the functions and focus on the business entities like:
- Location level
- Initiative level
- Process scorecard
- Product scorecard
TIP- Given so many scorecards being generated for so many reasons, a health check on the consistency of data helps. This health check is inherently done, when the lower level scorecards roll-up into higher level. However, it becomes an issue, if the scorecards are not inter-related. One way to get this done is that if we pick data for the scorecards from enterprise reports, as much as possible. We do expect and hope that reports are consistent in data terms.
Scorecard Generation is not an Additional Task
People dread creating and reviewing scorecards. The first reason is that they consider it as an additional task which can be dispensed with. Secondly, they want to avoid seeing the current state in the face and also to project their future performance.
In the opinion of ExecutionMiH.com, Scorecard is a way of life and style of management. Everything which is done in scorecard generation would have been done irrespective by a prudent management. Scorecards give a shape and discipline to these prudent practices. The challenge is not with the process, but with the efficiency of the process.
|
