|
What KPIs to keep and what not to keep? How many KPIs per user?, How many KPIs per dashboard?, How many KPIs at what level?- Are few of the question among dozens faced by the performance managers. KPIs/Metrics are used both in Dashboard and scorecards. Here we are going to share some of the rules on what could be an appropriate approach.
High frequency reviews should look for simpler KPIs
More frequently you are reviewing a KPI, simpler it needs to be. The reason is simple. A complex KPI takes more time to analyze. By the time you analyze it and decide about the action you need to take (improve or sustain or reduce), you may not have enough time to act, before the next review cycle starts. For example, if you have the sales revenue per office or sales pipelines per office, its a good and simple metric. However, if you place the average sales productivity on your daily dashboard, it will be difficult to manage it as the analysis required to find the root causes of the sales productivity issues will be time consuming.
At the same time, when you talk about a scorecard (monthly or fortnightly), it can have more complex KPIs, as one has time to analyze and take action around the KPI till the next cycle.
Higher is the review frequency, lesser should be the KPIs
This comes out of the concept of the Dashboard vs. scorecards. A dashboard, for example is generated at high frequency (daily basis). You will have limited number of simple KPIs. If you have got shorter review cycles, you will need to focus on the limited number of KPIs, which you can handle in terms of analysis and response. If a daily dashboard has 10 (say) KPIs, chances are that not all of them will be worked-upon. Scorecards on the other-hand can have many more KPIs.
5-7 KPIs in a given Level
This is more related to the scorecards. A scorecard (because of it monthly frequency) can have complex and large number of KPIs. However, again going as per a human's comprehension and view-window capabilities, one should limit the number of KPIs for a given level of a scorecard. For example, one can have the sales productivity at the top level and then the other contributing KPIs (say sales velocity, sales strike rate...) at the next level. The principle applied is that one should not have overlapping KPIs, and high priority KPIs.
Keep the exception KPIs
Sometimes the priority of a KPI is decided not only by its alignment with the business goals, but also on the levels of exception. For example you may have the order fulfillment TAT to be core to the business goals, but you may decide not to keep it in KPIs, as you have been performing well on these KPIs and beating the expected standard. This does not mean that this KPI will not be on your radar, but it may not be the 'front-page news' on your scorecard. At the same time, you would place higher emphasis on the KPIs, which are undergoing frequent exceptions.
For scorecards, place greater emphasis on the KPIs, which have not been reviewed in the dashboards
If I have been keeping a weekly (or daily) watch on the orders booked/expenses made/deliveries done/revenue realized etc., I will not need to spend too much space on these frequently reviewed metrics in my monthly scorecards. The frequent review items would have their story worked out through the month and their appearance in the monthly scorecard will be more of FYI. We would recommend to spend more time and space on the areas, which have not been reviewed through more frequent dashboards. For examples while revenue related metrics are review through dashboards, the sales productivity may be reviewed on monthly basis in great detail.
Do not prioritize on the basis of 'degree of confidence' of KPI data
Sometimes, you have KPIs, which are little qualitative (for example customer satisfaction scores) or may not be accurate in terms of measurement (for example the TAT for responding to over the counter submission of application by the agent.), as the data comes from loose systems or the excel sheets. However, the prioritization for a KPI (in terms of the time-space it will get in the scorecard), should not be influenced by this lack of accuracy or quantification. However, if the level of confidence on the KPI data is below a given thresh-hold, it will be advisable to not to include it.
|
