We don’t know what a KPI, an OKR or a metric is. We like the acronyms and use them because many years ago we heard them from a guy in a tie along with the words “synergy” and “quick win”.
We apply these concepts interchangeably because they have parts in common. Let’s learn to tell them apart.
A metric is a system or standard of measurement. In the technology world it is usually defined as a set of numbers or statistics that measure results. For example: monthly active users, Net Promoter Score Customer Renewal Rate, EBIT, Casflow, burn rate.
A metric always measures a current value, i.e. the value of the metric at this moment. A metric is neither a KPI nor an OKR, even if they are used in both. A KPI or an OKR uses metrics, but they are not just metrics.
A KPI consists of a metric, a current value, a target value and a title.
An OKR, or main objective indicator, always pertains to an objective, which is no surprise with that name. It kills two birds with one stone, avoids ambiguity by understanding what the value is that defines success and helps measure progress to the target. It presents an additional value, the starting value within a measurement timeline.
A good OKR, would be “Increase NPS from 20 to 40 by Q3 2022”, where the metric would be NPS, the start value 20 and the target value 40.
A KPI is a tool to monitor how an area under analysis behaves. It always has a metric to measure performance in that area.
An OKR is a tool that combines the target with the current state, desired state and starting point on the timeline under analysis.
And now, forget what you’ve read, but remember that KPIs and OKRs use metrics, but they are more than just metrics.
When metrics are designed as the basis for a company’s KPIs or OKRs aligned with objectives, you can be sure that they will be achieved, often without even thinking about whether or not they make sense. KPIs and OKRs make us see the map, but not the road.
KPIs or OKRs are defined after a strategic analysis and design usually at the beginning of the year, where competent agilists or anyone with sufficient motor skills move colored papers on a board that has a powerful name like “Balanced Scorecard”, just like Hitler in the Chancellery bunker moved imaginary divisions.
This is usually done after a SWOT analysis, crossing combinations, looking each other in the eye and drawing matrices to predict the behavior of markets and scenarios with the global economy in the background and the Ukrainian crisis as a musical thread.
Now the metrics, KPIs and OKRs, and their timeframes, are already taken out and handed out to people who haven’t been to any meetings and don’t even know what they are talking about. So everyone is busy making up forms whose relationship with reality is anecdotal.
Then KPIs and OKRs are reviewed to see if they are “aligned”, God knows what that means… and we prepare for the end of the year, where employees grow if the company grows, if the global economy grows, in a context, remember, where we have the Russians….
Objectives must be adaptive and accept context changes. Flexible, assuming that strategies represent a will, without a necessary relationship with a can, that objectives and metrics must be nuanced, refined and brought closer to reality and not mistreating reality to bring it closer to the former.
Annual milestones are predefined where commitments are reviewed and the strategy is adapted, but there is no obligation to follow an immutable established plan. In addition, this Agile approach must emphasize values.
Otherwise it’s all very perverse, you’ll see. Imagine that we value the efficiency of a police station by the increase in arrests. If you read it quickly, it makes sense, but if you analyze it, having more arrests does not mean getting more criminals off the street, or even that the justice process can convict them.
Detention metrics directly impact the value of justice. Values should be the guiding lines for goal setting.
Data are valuable depending on how they are used, but they are not valuable for predictions, only trends. Lessons learned from past data and the consequences learned from them are irrelevant. After the 9/11 attacks, all efforts were directed at preventing the conditions that led to the attacks. But those conditions had already occurred and were not going to happen again, there could be new attacks, but they would be executed in a different way, for example by stabbing in the main Western metropolises.
Adaptation implies that we will change the targets if it is shown that others are better suited to the strategy and the circumstances during the year, to the clients, to the people. There is no point in arresting a lot of criminals if you don’t manage to increase the security of a society.