At the beginning of our work with a client, we often bring together a team of staff including executives and managers to construct a framework which describes the mission of the organization and the priority intermediate outcomes which combine to producing that impact.  It's an interesting process as a wide range of interests, needs, and responsibilities perspectives come together to form a cohesive whole.  It not only provides a context for ongoing strategic conversations, but it also helps each person see the value of their worth in the context of the organization.

But as our work shifts from framework development to defining the metrics which indicate progress for key outcomes, the breadth of interests across the organization resurfaces.  Suddenly, we've created a measurement hammer and every data need starts looking like a nail.  But each level of management has its own set of needs.

The Board needs to know if the organization is accomplishing its mission.  Senior leadership needs to know how well the key intermediate outcomes are being advanced so it can allocate resources accordingly.  Management needs know how those outcomes are being produced and where in the underlying activities or decisions problems lie so it can use its resources more effectively.  And on the ground, all staff need access to the data which helps them to their jobs most effectively.

The type and quantity of data needed to addresses each audience differs significantly.  In an ideal world, there's a single data system in which all data is tracked and through which all reports are produced. Everything simply rolls up together and can be sliced and diced in any way at the drop of a hat.  It's a great concept.  And while versions of that ideal world exist in corners of organizations, I've yet to see it in practice across the board for any number of reasons.  There are different systems on the ground level.  People define things a little bit differently in different departments.  The data isn't captured reliably (or at all).  Technical staff, if the organization has dedicated technical staff, can't keep up with all the requests.  These sorts of challenges exist throughout all organizations.

Each level of needs which you are addressing will carry with it a new set of data needs.  Any measurement effort which tries to tackle everything at once is bound to be overwhelmed.  While all of the needs listed above matter, what matters most?  What decision or report, if properly informed by measurement, which have the most impact on your organization?  Answer this question first and then start simple and stay focused. Define the need clearly and outline how the data will be used or reported. Use existing data and data collection processes where possible; create simple new processes and collect new data where necessary.  Only collect the data you need.  Over time, as your capacity to measure expands, you can expand your efforts the range of needs you address accordingly.

As important as it is to address needs across an organization, it is important to demonstrate that you can do it well.  By focusing first on a single level of data and executing well, you lay a strong foundation for being a data-driven organization in to the future.