Insights from large sets of data are not always found in a straight line. In our domain—analyzing data from time utilization studies—it often takes the spark of curiosity from one set of results to ignite the question that delivers real value.
Because the data that we gather for time utilization studies has a consistent structure, we have over 40 standard reports that fall into three groups: occupancy utilization, primary activity, and number of people in the room. For each of these groups, we can instantly show all-up averages, or as broken out by room type, or by day of the week, or by hour of the day. Pivot a few of these, and Voila! Over 40 reports can be run even as the data is still being gathered. Sounds pretty substantial, right?
Then we got a client who pushed us to go deeper.
“John” asked us about vacancy. Not just percent of occupancy, but verifiable, specific full-day vacancy. And once we gave him that, he asked for half-day vacancy. The point was that he needed to show hard data to specific departments about how their spaces were being used, by role. It wasn’t as straightforward as it seems; we had to run queries off of the results of other queries. We ended up with a 3-phased analysis to reveal the full and half-day vacancies.
For a full description of this process, you can download our case study here.
Here’s a breakdown of the complexity. For a general understanding of “under-utilized,” an empty office could be recorded as “unoccupied” on any one of the nine passes recorded by the surveyor each day. However, this client wanted to know with a high degree of certainty that the occupant of the room was not present at any time during the day in order to count the room as vacant.
As an aside, the collected data already distinguished between “unoccupied” and “temporarily unoccupied.” Observers are always given detailed instructions to look for clues that an office occupant might just be away at lunch or at a meeting. A status of “unoccupied” would be determined by observing no coats on chairs, no snacks on the desk, or other indications that the employee had been at his/her desk that day. This distinction made a strong case for vacancy when full days of “unoccupied” were identified in the reporting.
So our three analytic phases broke down to:
- Phase 1: counting observations where primary activity is unoccupied
- Phase 2: Finding (filtering for) only the spaces that were unoccupied for a full day (unoccupied all nine passes in a day)
- Phase 3: Counting the full day vacancies per business unit by role
THEN, the half day vacancies were a twist on the three phases because they needed to first filter OUT all the full day vacancies as well as make sure that the partial vacancies had enough consecutive passes of “unoccupied” during the periods defined as half days. Not quite straightforward, right!
In sharing this report with another customer, a workplace strategist commented, “Interesting, and generally I look for the inverse, to find and highlight the highly utilized spaces.” Yes! It makes sense that one goal of a utilization study is to look for spaces that can immediately be re-purposed (the highly vacant), and another goal is to learn what to design more of into a new space (the highly utilized).
What insights have you found (or would like to find) in utilization data?